Hi
I hit the following 2 warnings when running with LOCKDEP=y on arm64 platform
(juno-r2), running on v5.6-rc6
The 1st one is when I execute `swapon -a`. The 2nd one happens at boot. I have
/dev/sda2 as my swap in /etc/fstab
Note that I either hit 1 OR 2, but didn't hit both warnings at the same time,
yet at least.
/dev/sda2 is a usb flash drive, in case it matters somehow.
### WARNING 1 ###
# swapon -a
[ 1682.788988]
[ 1682.790503] =====================================
[ 1682.795226] WARNING: bad unlock balance detected!
[ 1682.799951] 5.6.0-rc6 #533 Not tainted
[ 1682.803714] -------------------------------------
[ 1682.808437] swapon/437 is trying to release lock (&sb->s_type->i_mutex_key) at:
[ 1682.815797] [<ffff80001030c07c>] __arm64_sys_swapon+0x3fc/0x1140
[ 1682.821828] but there are no more locks to release!
[ 1682.826724]
[ 1682.826724] other info that might help us debug this:
[ 1682.833280] no locks held by swapon/437.
[ 1682.837216]
[ 1682.837216] stack backtrace:
[ 1682.841596] CPU: 4 PID: 437 Comm: swapon Not tainted 5.6.0-rc6 #533
[ 1682.847889] Hardware name: ARM Juno development board (r2) (DT)
[ 1682.853834] Call trace:
[ 1682.856293] dump_backtrace+0x0/0x1a8
[ 1682.859971] show_stack+0x24/0x30
[ 1682.863302] dump_stack+0xe8/0x150
[ 1682.866722] print_unlock_imbalance_bug+0xe4/0xe8
[ 1682.871447] lock_release+0x274/0x350
[ 1682.875128] up_write+0x30/0x190
[ 1682.878371] __arm64_sys_swapon+0x3fc/0x1140
[ 1682.882662] el0_svc_common.constprop.2+0x7c/0x178
[ 1682.887475] do_el0_svc+0x34/0xa0
[ 1682.890805] el0_sync_handler+0x114/0x1d0
[ 1682.894832] el0_sync+0x140/0x180
[ 1682.898233] ------------[ cut here ]------------
[ 1682.902936] DEBUG_RWSEMS_WARN_ON((rwsem_owner(sem) != current) && !rwsem_test_oflags(sem, RWSEM_NONSPINNABLE)): count = 0x0, magic = 0xffff000971018270, owner = 0x0, curr 0xffff0009752fc680, list empty
[ 1682.921045] WARNING: CPU: 4 PID: 437 at kernel/locking/rwsem.c:1459 up_write+0x134/0x190
[ 1682.929203] Modules linked in:
[ 1682.932288] CPU: 4 PID: 437 Comm: swapon Not tainted 5.6.0-rc6 #533
[ 1682.938604] Hardware name: ARM Juno development board (r2) (DT)
[ 1682.944574] pstate: 60000005 (nZCv daif -PAN -UAO)
[ 1682.949407] pc : up_write+0x134/0x190
[ 1682.953102] lr : up_write+0x134/0x190
[ 1682.956793] sp : ffff800018e1bd30
[ 1682.960135] x29: ffff800018e1bd30 x28: ffff000970fb4000
[ 1682.965494] x27: ffff0009750bd000 x26: 00000000fffffff0
[ 1682.970852] x25: 0000000000000000 x24: fffffffffffffff0
[ 1682.976211] x23: ffff000971439200 x22: ffff80001323d000
[ 1682.981569] x21: ffff800013241000 x20: ffff80001030c07c
[ 1682.986927] x19: ffff000971018270 x18: ffffffffffffffff
[ 1682.992285] x17: 0000000000000000 x16: 0000000000000000
[ 1682.997643] x15: ffff80001323da88 x14: 0720072007200720
[ 1683.003002] x13: 0720072007200720 x12: 0720072007200720
[ 1683.008360] x11: 0720072007200720 x10: 0000000000000000
[ 1683.013718] x9 : ffff80001323da88 x8 : ffff800013267808
[ 1683.019076] x7 : ffff80001019f77c x6 : ffff800018e1b9f0
[ 1683.024434] x5 : 0000000000000001 x4 : 0000000000000001
[ 1683.029791] x3 : 775f041afc1e3200 x2 : ffff800013267860
[ 1683.035149] x1 : 775f041afc1e3200 x0 : 0000000000000000
[ 1683.040507] Call trace:
[ 1683.042977] up_write+0x134/0x190
[ 1683.046323] __arm64_sys_swapon+0x3fc/0x1140
[ 1683.050632] el0_svc_common.constprop.2+0x7c/0x178
[ 1683.055464] do_el0_svc+0x34/0xa0
[ 1683.058810] el0_sync_handler+0x114/0x1d0
[ 1683.062854] el0_sync+0x140/0x180
[ 1683.066196] irq event stamp: 1095
[ 1683.069543] hardirqs last enabled at (1095): [<ffff80001191b724>] _raw_spin_unlock_irqrestore+0x7c/0x88
[ 1683.079104] hardirqs last disabled at (1094): [<ffff80001191b4c0>] _raw_spin_lock_irqsave+0x38/0x88
[ 1683.088226] softirqs last enabled at (996): [<ffff8000100818a4>] __do_softirq+0x4bc/0x568
[ 1683.096562] softirqs last disabled at (985): [<ffff800010114064>] irq_exit+0x144/0x150
[ 1683.104545] ---[ end trace 737bf4f253f20ead ]---
### WARNING 2 ###
[ 15.502583] Unable to find swap-space signature
[ 15.507496]
[ 15.508996] ======================================================
[ 15.515202] WARNING: possible circular locking dependency detected
[ 15.521411] 5.6.0-rc6 #533 Not tainted
[ 15.525173] ------------------------------------------------------
[ 15.531378] swapon/321 is trying to acquire lock:
[ 15.536102] ffff000971020078 (&bdev->bd_mutex){+.+.}, at: blkdev_put+0x30/0x120
[ 15.543460]
[ 15.543460] but task is already holding lock:
[ 15.549317] ffff0009710202d8 (&sb->s_type->i_mutex_key#7){+.+.}, at: __arm64_sys_swapon+0x2cc/0x1140
[ 15.558505]
[ 15.558505] which lock already depends on the new lock.
[ 15.558505]
[ 15.566719]
[ 15.566719] the existing dependency chain (in reverse order) is:
[ 15.574234]
[ 15.574234] -> #1 (&sb->s_type->i_mutex_key#7){+.+.}:
[ 15.580802] down_write+0x50/0xb0
[ 15.584656] __blkdev_get+0x2a4/0x4f8
[ 15.588859] blkdev_get+0xfc/0x190
[ 15.592800] __device_add_disk+0x470/0x4b8
[ 15.597437] device_add_disk+0x38/0x48
[ 15.601729] sd_probe+0x350/0x488
[ 15.605585] really_probe+0x10c/0x358
[ 15.609788] driver_probe_device+0x5c/0x100
[ 15.614514] __device_attach_driver+0x98/0xb8
[ 15.619415] bus_for_each_drv+0x74/0xd8
[ 15.623792] __device_attach_async_helper+0xc8/0xf0
[ 15.629217] async_run_entry_fn+0x48/0x148
[ 15.633857] process_one_work+0x2a4/0x748
[ 15.638407] worker_thread+0x48/0x498
[ 15.642609] kthread+0x13c/0x140
[ 15.646376] ret_from_fork+0x10/0x18
[ 15.650487]
[ 15.650487] -> #0 (&bdev->bd_mutex){+.+.}:
[ 15.656093] __lock_acquire+0x1008/0x1450
[ 15.660645] lock_acquire+0xf8/0x280
[ 15.664759] __mutex_lock+0xac/0x920
[ 15.668873] mutex_lock_nested+0x3c/0x50
[ 15.673337] blkdev_put+0x30/0x120
[ 15.677278] __arm64_sys_swapon+0x5a8/0x1140
[ 15.682093] el0_svc_common.constprop.2+0x7c/0x178
[ 15.687429] do_el0_svc+0x34/0xa0
[ 15.691282] el0_sync_handler+0x114/0x1d0
[ 15.695833] el0_sync+0x140/0x180
[ 15.699683]
[ 15.699683] other info that might help us debug this:
[ 15.699683]
[ 15.707721] Possible unsafe locking scenario:
[ 15.707721]
[ 15.713665] CPU0 CPU1
[ 15.718213] ---- ----
[ 15.722759] lock(&sb->s_type->i_mutex_key#7);
[ 15.727311] lock(&bdev->bd_mutex);
[ 15.733430] lock(&sb->s_type->i_mutex_key#7);
[ 15.740510] lock(&bdev->bd_mutex);
[ 15.744100]
[ 15.744100] *** DEADLOCK ***
[ 15.744100]
[ 15.750048] 1 lock held by swapon/321:
[ 15.753810] #0: ffff0009710202d8 (&sb->s_type->i_mutex_key#7){+.+.}, at: __arm64_sys_swapon+0x2cc/0x1140
[ 15.763432]
[ 15.763432] stack backtrace:
[ 15.767811] CPU: 3 PID: 321 Comm: swapon Not tainted 5.6.0-rc6 #533
[ 15.774104] Hardware name: ARM Juno development board (r2) (DT)
[ 15.780048] Call trace:
[ 15.782505] dump_backtrace+0x0/0x1a8
[ 15.786182] show_stack+0x24/0x30
[ 15.789512] dump_stack+0xe8/0x150
[ 15.792928] print_circular_bug.isra.39+0x1b8/0x210
[ 15.797826] check_noncircular+0x178/0x1e0
[ 15.801940] __lock_acquire+0x1008/0x1450
[ 15.805967] lock_acquire+0xf8/0x280
[ 15.809558] __mutex_lock+0xac/0x920
[ 15.813149] mutex_lock_nested+0x3c/0x50
[ 15.817089] blkdev_put+0x30/0x120
[ 15.820507] __arm64_sys_swapon+0x5a8/0x1140
[ 15.824797] el0_svc_common.constprop.2+0x7c/0x178
[ 15.829609] do_el0_svc+0x34/0xa0
[ 15.832938] el0_sync_handler+0x114/0x1d0
[ 15.836964] el0_sync+0x140/0x180
swapon: /dev/sda2: Invalid argument
Thanks
--
Qais Yousef
On 03/23/20 15:17, Qais Yousef wrote:
> Hi
>
> I hit the following 2 warnings when running with LOCKDEP=y on arm64 platform
> (juno-r2), running on v5.6-rc6
>
> The 1st one is when I execute `swapon -a`. The 2nd one happens at boot. I have
> /dev/sda2 as my swap in /etc/fstab
>
> Note that I either hit 1 OR 2, but didn't hit both warnings at the same time,
> yet at least.
>
> /dev/sda2 is a usb flash drive, in case it matters somehow.
By the way, I noticed that in claim_swapfile() if we fail we don't release the
lock. Shouldn't we release the lock here?
I tried with that FWIW, but it had no effect on the warnings.
diff --git a/mm/swapfile.c b/mm/swapfile.c
index b2a2e45c9a36..0293e8c7b472 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -2900,8 +2900,10 @@ static int claim_swapfile(struct swap_info_struct *p, struct inode *inode)
}
inode_lock(inode);
- if (IS_SWAPFILE(inode))
+ if (IS_SWAPFILE(inode)) {
+ inode_unlock(inode);
return -EBUSY;
+ }
return 0;
}
Thanks
--
Qais Yousef
On Mon, 23 Mar 2020 15:30:45 +0000 Qais Yousef <[email protected]> wrote:
> On 03/23/20 15:17, Qais Yousef wrote:
> > Hi
> >
> > I hit the following 2 warnings when running with LOCKDEP=y on arm64 platform
> > (juno-r2), running on v5.6-rc6
> >
> > The 1st one is when I execute `swapon -a`. The 2nd one happens at boot. I have
> > /dev/sda2 as my swap in /etc/fstab
> >
> > Note that I either hit 1 OR 2, but didn't hit both warnings at the same time,
> > yet at least.
> >
> > /dev/sda2 is a usb flash drive, in case it matters somehow.
>
> By the way, I noticed that in claim_swapfile() if we fail we don't release the
> lock. Shouldn't we release the lock here?
>
> I tried with that FWIW, but it had no effect on the warnings.
>
I'll be sending the below into Linus this week.
I was hoping to hear from Darrick/Christoph (?) but it looks like the
right thing to do. Are you able to test it?
I think I'll add a cc:stable to this one.
From: Naohiro Aota <[email protected]>
Subject: mm/swapfile.c: move inode_lock out of claim_swapfile
claim_swapfile() currently keeps the inode locked when it is successful,
or the file is already swapfile (with -EBUSY). And, on the other error
cases, it does not lock the inode.
This inconsistency of the lock state and return value is quite confusing
and actually causing a bad unlock balance as below in the "bad_swap"
section of __do_sys_swapon().
This commit fixes this issue by moving the inode_lock() and IS_SWAPFILE
check out of claim_swapfile(). The inode is unlocked in
"bad_swap_unlock_inode" section, so that the inode is ensured to be
unlocked at "bad_swap". Thus, error handling codes after the locking now
jumps to "bad_swap_unlock_inode" instead of "bad_swap".
=====================================
WARNING: bad unlock balance detected!
5.5.0-rc7+ #176 Not tainted
-------------------------------------
swapon/4294 is trying to release lock (&sb->s_type->i_mutex_key) at:
[<ffffffff8173a6eb>] __do_sys_swapon+0x94b/0x3550
but there are no more locks to release!
other info that might help us debug this:
no locks held by swapon/4294.
stack backtrace:
CPU: 5 PID: 4294 Comm: swapon Not tainted 5.5.0-rc7-BTRFS-ZNS+ #176
Hardware name: ASUS All Series/H87-PRO, BIOS 2102 07/29/2014
Call Trace:
dump_stack+0xa1/0xea
? __do_sys_swapon+0x94b/0x3550
print_unlock_imbalance_bug.cold+0x114/0x123
? __do_sys_swapon+0x94b/0x3550
lock_release+0x562/0xed0
? kvfree+0x31/0x40
? lock_downgrade+0x770/0x770
? kvfree+0x31/0x40
? rcu_read_lock_sched_held+0xa1/0xd0
? rcu_read_lock_bh_held+0xb0/0xb0
up_write+0x2d/0x490
? kfree+0x293/0x2f0
__do_sys_swapon+0x94b/0x3550
? putname+0xb0/0xf0
? kmem_cache_free+0x2e7/0x370
? do_sys_open+0x184/0x3e0
? generic_max_swapfile_size+0x40/0x40
? do_syscall_64+0x27/0x4b0
? entry_SYSCALL_64_after_hwframe+0x49/0xbe
? lockdep_hardirqs_on+0x38c/0x590
__x64_sys_swapon+0x54/0x80
do_syscall_64+0xa4/0x4b0
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x7f15da0a0dc7
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 1638045c3677 ("mm: set S_SWAPFILE on blockdev swap devices")
Signed-off-by: Naohiro Aota <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Darrick J. Wong <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
---
mm/swapfile.c | 41 ++++++++++++++++++++---------------------
1 file changed, 20 insertions(+), 21 deletions(-)
--- a/mm/swapfile.c~mm-swap-move-inode_lock-out-of-claim_swapfile
+++ a/mm/swapfile.c
@@ -2899,10 +2899,6 @@ static int claim_swapfile(struct swap_in
p->bdev = inode->i_sb->s_bdev;
}
- inode_lock(inode);
- if (IS_SWAPFILE(inode))
- return -EBUSY;
-
return 0;
}
@@ -3157,36 +3153,41 @@ SYSCALL_DEFINE2(swapon, const char __use
mapping = swap_file->f_mapping;
inode = mapping->host;
- /* will take i_rwsem; */
error = claim_swapfile(p, inode);
if (unlikely(error))
goto bad_swap;
+ inode_lock(inode);
+ if (IS_SWAPFILE(inode)) {
+ error = -EBUSY;
+ goto bad_swap_unlock_inode;
+ }
+
/*
* Read the swap header.
*/
if (!mapping->a_ops->readpage) {
error = -EINVAL;
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
page = read_mapping_page(mapping, 0, swap_file);
if (IS_ERR(page)) {
error = PTR_ERR(page);
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
swap_header = kmap(page);
maxpages = read_swap_header(p, swap_header, inode);
if (unlikely(!maxpages)) {
error = -EINVAL;
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
/* OK, set up the swap map and apply the bad block list */
swap_map = vzalloc(maxpages);
if (!swap_map) {
error = -ENOMEM;
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
if (bdi_cap_stable_pages_required(inode_to_bdi(inode)))
@@ -3211,7 +3212,7 @@ SYSCALL_DEFINE2(swapon, const char __use
GFP_KERNEL);
if (!cluster_info) {
error = -ENOMEM;
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
for (ci = 0; ci < nr_cluster; ci++)
@@ -3220,7 +3221,7 @@ SYSCALL_DEFINE2(swapon, const char __use
p->percpu_cluster = alloc_percpu(struct percpu_cluster);
if (!p->percpu_cluster) {
error = -ENOMEM;
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
for_each_possible_cpu(cpu) {
struct percpu_cluster *cluster;
@@ -3234,13 +3235,13 @@ SYSCALL_DEFINE2(swapon, const char __use
error = swap_cgroup_swapon(p->type, maxpages);
if (error)
- goto bad_swap;
+ goto bad_swap_unlock_inode;
nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map,
cluster_info, maxpages, &span);
if (unlikely(nr_extents < 0)) {
error = nr_extents;
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
/* frontswap enabled? set up bit-per-page map for frontswap */
if (IS_ENABLED(CONFIG_FRONTSWAP))
@@ -3280,7 +3281,7 @@ SYSCALL_DEFINE2(swapon, const char __use
error = init_swap_address_space(p->type, maxpages);
if (error)
- goto bad_swap;
+ goto bad_swap_unlock_inode;
/*
* Flush any pending IO and dirty mappings before we start using this
@@ -3290,7 +3291,7 @@ SYSCALL_DEFINE2(swapon, const char __use
error = inode_drain_writes(inode);
if (error) {
inode->i_flags &= ~S_SWAPFILE;
- goto bad_swap;
+ goto bad_swap_unlock_inode;
}
mutex_lock(&swapon_mutex);
@@ -3315,6 +3316,8 @@ SYSCALL_DEFINE2(swapon, const char __use
error = 0;
goto out;
+bad_swap_unlock_inode:
+ inode_unlock(inode);
bad_swap:
free_percpu(p->percpu_cluster);
p->percpu_cluster = NULL;
@@ -3322,6 +3325,7 @@ bad_swap:
set_blocksize(p->bdev, p->old_block_size);
blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
}
+ inode = NULL;
destroy_swap_extents(p);
swap_cgroup_swapoff(p->type);
spin_lock(&swap_lock);
@@ -3333,13 +3337,8 @@ bad_swap:
kvfree(frontswap_map);
if (inced_nr_rotate_swap)
atomic_dec(&nr_rotate_swap);
- if (swap_file) {
- if (inode) {
- inode_unlock(inode);
- inode = NULL;
- }
+ if (swap_file)
filp_close(swap_file, NULL);
- }
out:
if (page && !IS_ERR(page)) {
kunmap(page);
_
On Mon, Mar 23, 2020 at 02:06:15PM -0700, Andrew Morton wrote:
> On Mon, 23 Mar 2020 15:30:45 +0000 Qais Yousef <[email protected]> wrote:
>
> > On 03/23/20 15:17, Qais Yousef wrote:
> > > Hi
> > >
> > > I hit the following 2 warnings when running with LOCKDEP=y on arm64 platform
> > > (juno-r2), running on v5.6-rc6
> > >
> > > The 1st one is when I execute `swapon -a`. The 2nd one happens at boot. I have
> > > /dev/sda2 as my swap in /etc/fstab
> > >
> > > Note that I either hit 1 OR 2, but didn't hit both warnings at the same time,
> > > yet at least.
> > >
> > > /dev/sda2 is a usb flash drive, in case it matters somehow.
> >
> > By the way, I noticed that in claim_swapfile() if we fail we don't release the
> > lock. Shouldn't we release the lock here?
> >
> > I tried with that FWIW, but it had no effect on the warnings.
> >
>
> I'll be sending the below into Linus this week.
>
> I was hoping to hear from Darrick/Christoph (?) but it looks like the
> right thing to do. Are you able to test it?
I had questions[1] about the patch, but nobody ever replied.
--D
[1] https://lore.kernel.org/linux-fsdevel/20200212164956.GK6874@magnolia/
> I think I'll add a cc:stable to this one.
>
>
>
> From: Naohiro Aota <[email protected]>
> Subject: mm/swapfile.c: move inode_lock out of claim_swapfile
>
> claim_swapfile() currently keeps the inode locked when it is successful,
> or the file is already swapfile (with -EBUSY). And, on the other error
> cases, it does not lock the inode.
>
> This inconsistency of the lock state and return value is quite confusing
> and actually causing a bad unlock balance as below in the "bad_swap"
> section of __do_sys_swapon().
>
> This commit fixes this issue by moving the inode_lock() and IS_SWAPFILE
> check out of claim_swapfile(). The inode is unlocked in
> "bad_swap_unlock_inode" section, so that the inode is ensured to be
> unlocked at "bad_swap". Thus, error handling codes after the locking now
> jumps to "bad_swap_unlock_inode" instead of "bad_swap".
>
> =====================================
> WARNING: bad unlock balance detected!
> 5.5.0-rc7+ #176 Not tainted
> -------------------------------------
> swapon/4294 is trying to release lock (&sb->s_type->i_mutex_key) at:
> [<ffffffff8173a6eb>] __do_sys_swapon+0x94b/0x3550
> but there are no more locks to release!
>
> other info that might help us debug this:
> no locks held by swapon/4294.
>
> stack backtrace:
> CPU: 5 PID: 4294 Comm: swapon Not tainted 5.5.0-rc7-BTRFS-ZNS+ #176
> Hardware name: ASUS All Series/H87-PRO, BIOS 2102 07/29/2014
> Call Trace:
> dump_stack+0xa1/0xea
> ? __do_sys_swapon+0x94b/0x3550
> print_unlock_imbalance_bug.cold+0x114/0x123
> ? __do_sys_swapon+0x94b/0x3550
> lock_release+0x562/0xed0
> ? kvfree+0x31/0x40
> ? lock_downgrade+0x770/0x770
> ? kvfree+0x31/0x40
> ? rcu_read_lock_sched_held+0xa1/0xd0
> ? rcu_read_lock_bh_held+0xb0/0xb0
> up_write+0x2d/0x490
> ? kfree+0x293/0x2f0
> __do_sys_swapon+0x94b/0x3550
> ? putname+0xb0/0xf0
> ? kmem_cache_free+0x2e7/0x370
> ? do_sys_open+0x184/0x3e0
> ? generic_max_swapfile_size+0x40/0x40
> ? do_syscall_64+0x27/0x4b0
> ? entry_SYSCALL_64_after_hwframe+0x49/0xbe
> ? lockdep_hardirqs_on+0x38c/0x590
> __x64_sys_swapon+0x54/0x80
> do_syscall_64+0xa4/0x4b0
> entry_SYSCALL_64_after_hwframe+0x49/0xbe
> RIP: 0033:0x7f15da0a0dc7
>
> Link: http://lkml.kernel.org/r/[email protected]
> Fixes: 1638045c3677 ("mm: set S_SWAPFILE on blockdev swap devices")
> Signed-off-by: Naohiro Aota <[email protected]>
> Reviewed-by: Andrew Morton <[email protected]>
> Cc: Darrick J. Wong <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> Signed-off-by: Andrew Morton <[email protected]>
> ---
>
> mm/swapfile.c | 41 ++++++++++++++++++++---------------------
> 1 file changed, 20 insertions(+), 21 deletions(-)
>
> --- a/mm/swapfile.c~mm-swap-move-inode_lock-out-of-claim_swapfile
> +++ a/mm/swapfile.c
> @@ -2899,10 +2899,6 @@ static int claim_swapfile(struct swap_in
> p->bdev = inode->i_sb->s_bdev;
> }
>
> - inode_lock(inode);
> - if (IS_SWAPFILE(inode))
> - return -EBUSY;
> -
> return 0;
> }
>
> @@ -3157,36 +3153,41 @@ SYSCALL_DEFINE2(swapon, const char __use
> mapping = swap_file->f_mapping;
> inode = mapping->host;
>
> - /* will take i_rwsem; */
> error = claim_swapfile(p, inode);
> if (unlikely(error))
> goto bad_swap;
>
> + inode_lock(inode);
> + if (IS_SWAPFILE(inode)) {
> + error = -EBUSY;
> + goto bad_swap_unlock_inode;
> + }
> +
> /*
> * Read the swap header.
> */
> if (!mapping->a_ops->readpage) {
> error = -EINVAL;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> page = read_mapping_page(mapping, 0, swap_file);
> if (IS_ERR(page)) {
> error = PTR_ERR(page);
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> swap_header = kmap(page);
>
> maxpages = read_swap_header(p, swap_header, inode);
> if (unlikely(!maxpages)) {
> error = -EINVAL;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> /* OK, set up the swap map and apply the bad block list */
> swap_map = vzalloc(maxpages);
> if (!swap_map) {
> error = -ENOMEM;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> if (bdi_cap_stable_pages_required(inode_to_bdi(inode)))
> @@ -3211,7 +3212,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> GFP_KERNEL);
> if (!cluster_info) {
> error = -ENOMEM;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> for (ci = 0; ci < nr_cluster; ci++)
> @@ -3220,7 +3221,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> p->percpu_cluster = alloc_percpu(struct percpu_cluster);
> if (!p->percpu_cluster) {
> error = -ENOMEM;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> for_each_possible_cpu(cpu) {
> struct percpu_cluster *cluster;
> @@ -3234,13 +3235,13 @@ SYSCALL_DEFINE2(swapon, const char __use
>
> error = swap_cgroup_swapon(p->type, maxpages);
> if (error)
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
>
> nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map,
> cluster_info, maxpages, &span);
> if (unlikely(nr_extents < 0)) {
> error = nr_extents;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> /* frontswap enabled? set up bit-per-page map for frontswap */
> if (IS_ENABLED(CONFIG_FRONTSWAP))
> @@ -3280,7 +3281,7 @@ SYSCALL_DEFINE2(swapon, const char __use
>
> error = init_swap_address_space(p->type, maxpages);
> if (error)
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
>
> /*
> * Flush any pending IO and dirty mappings before we start using this
> @@ -3290,7 +3291,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> error = inode_drain_writes(inode);
> if (error) {
> inode->i_flags &= ~S_SWAPFILE;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> mutex_lock(&swapon_mutex);
> @@ -3315,6 +3316,8 @@ SYSCALL_DEFINE2(swapon, const char __use
>
> error = 0;
> goto out;
> +bad_swap_unlock_inode:
> + inode_unlock(inode);
> bad_swap:
> free_percpu(p->percpu_cluster);
> p->percpu_cluster = NULL;
> @@ -3322,6 +3325,7 @@ bad_swap:
> set_blocksize(p->bdev, p->old_block_size);
> blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
> }
> + inode = NULL;
> destroy_swap_extents(p);
> swap_cgroup_swapoff(p->type);
> spin_lock(&swap_lock);
> @@ -3333,13 +3337,8 @@ bad_swap:
> kvfree(frontswap_map);
> if (inced_nr_rotate_swap)
> atomic_dec(&nr_rotate_swap);
> - if (swap_file) {
> - if (inode) {
> - inode_unlock(inode);
> - inode = NULL;
> - }
> + if (swap_file)
> filp_close(swap_file, NULL);
> - }
> out:
> if (page && !IS_ERR(page)) {
> kunmap(page);
> _
>
On Mon, Mar 23, 2020 at 02:12:01PM -0700, Darrick J. Wong wrote:
>On Mon, Mar 23, 2020 at 02:06:15PM -0700, Andrew Morton wrote:
>> On Mon, 23 Mar 2020 15:30:45 +0000 Qais Yousef <[email protected]> wrote:
>>
>> > On 03/23/20 15:17, Qais Yousef wrote:
>> > > Hi
>> > >
>> > > I hit the following 2 warnings when running with LOCKDEP=y on arm64 platform
>> > > (juno-r2), running on v5.6-rc6
>> > >
>> > > The 1st one is when I execute `swapon -a`. The 2nd one happens at boot. I have
>> > > /dev/sda2 as my swap in /etc/fstab
>> > >
>> > > Note that I either hit 1 OR 2, but didn't hit both warnings at the same time,
>> > > yet at least.
>> > >
>> > > /dev/sda2 is a usb flash drive, in case it matters somehow.
>> >
>> > By the way, I noticed that in claim_swapfile() if we fail we don't release the
>> > lock. Shouldn't we release the lock here?
>> >
>> > I tried with that FWIW, but it had no effect on the warnings.
>> >
>>
>> I'll be sending the below into Linus this week.
>>
>> I was hoping to hear from Darrick/Christoph (?) but it looks like the
>> right thing to do. Are you able to test it?
>
>I had questions[1] about the patch, but nobody ever replied.
Sorry, I have overlooked that one. I'm replying here.
> Sorry to wander in late, but I don't see how we unlock the inode in the
> success case. Before this patch, the "if (inode) inode_unlock(inode);"
> below out: would take care of this for both the success case and the
> bad_swap case, but now that's gone, and AFAICT after this patch we only
> unlock the inode when erroring out...
>
> > +bad_swap_unlock_inode:
> > + inode_unlock(inode);
>
> ...since we never goto bad_swap_unlock_inode when error == 0, correct?
Actually, this patch does not touch "if (inode) inode_unlock(inode);" below
the "out:" label, so we still have that "inode_unlock" after this patch and
it unlocks the inode in the success case.
>
>--D
>
>[1] https://lore.kernel.org/linux-fsdevel/20200212164956.GK6874@magnolia/
>
>> I think I'll add a cc:stable to this one.
>>
>>
>>
>> From: Naohiro Aota <[email protected]>
>> Subject: mm/swapfile.c: move inode_lock out of claim_swapfile
>>
>> claim_swapfile() currently keeps the inode locked when it is successful,
>> or the file is already swapfile (with -EBUSY). And, on the other error
>> cases, it does not lock the inode.
>>
>> This inconsistency of the lock state and return value is quite confusing
>> and actually causing a bad unlock balance as below in the "bad_swap"
>> section of __do_sys_swapon().
>>
>> This commit fixes this issue by moving the inode_lock() and IS_SWAPFILE
>> check out of claim_swapfile(). The inode is unlocked in
>> "bad_swap_unlock_inode" section, so that the inode is ensured to be
>> unlocked at "bad_swap". Thus, error handling codes after the locking now
>> jumps to "bad_swap_unlock_inode" instead of "bad_swap".
>>
>> =====================================
>> WARNING: bad unlock balance detected!
>> 5.5.0-rc7+ #176 Not tainted
>> -------------------------------------
>> swapon/4294 is trying to release lock (&sb->s_type->i_mutex_key) at:
>> [<ffffffff8173a6eb>] __do_sys_swapon+0x94b/0x3550
>> but there are no more locks to release!
>>
>> other info that might help us debug this:
>> no locks held by swapon/4294.
>>
>> stack backtrace:
>> CPU: 5 PID: 4294 Comm: swapon Not tainted 5.5.0-rc7-BTRFS-ZNS+ #176
>> Hardware name: ASUS All Series/H87-PRO, BIOS 2102 07/29/2014
>> Call Trace:
>> dump_stack+0xa1/0xea
>> ? __do_sys_swapon+0x94b/0x3550
>> print_unlock_imbalance_bug.cold+0x114/0x123
>> ? __do_sys_swapon+0x94b/0x3550
>> lock_release+0x562/0xed0
>> ? kvfree+0x31/0x40
>> ? lock_downgrade+0x770/0x770
>> ? kvfree+0x31/0x40
>> ? rcu_read_lock_sched_held+0xa1/0xd0
>> ? rcu_read_lock_bh_held+0xb0/0xb0
>> up_write+0x2d/0x490
>> ? kfree+0x293/0x2f0
>> __do_sys_swapon+0x94b/0x3550
>> ? putname+0xb0/0xf0
>> ? kmem_cache_free+0x2e7/0x370
>> ? do_sys_open+0x184/0x3e0
>> ? generic_max_swapfile_size+0x40/0x40
>> ? do_syscall_64+0x27/0x4b0
>> ? entry_SYSCALL_64_after_hwframe+0x49/0xbe
>> ? lockdep_hardirqs_on+0x38c/0x590
>> __x64_sys_swapon+0x54/0x80
>> do_syscall_64+0xa4/0x4b0
>> entry_SYSCALL_64_after_hwframe+0x49/0xbe
>> RIP: 0033:0x7f15da0a0dc7
>>
>> Link: http://lkml.kernel.org/r/[email protected]
>> Fixes: 1638045c3677 ("mm: set S_SWAPFILE on blockdev swap devices")
>> Signed-off-by: Naohiro Aota <[email protected]>
>> Reviewed-by: Andrew Morton <[email protected]>
>> Cc: Darrick J. Wong <[email protected]>
>> Cc: Christoph Hellwig <[email protected]>
>> Signed-off-by: Andrew Morton <[email protected]>
>> ---
>>
>> mm/swapfile.c | 41 ++++++++++++++++++++---------------------
>> 1 file changed, 20 insertions(+), 21 deletions(-)
>>
>> --- a/mm/swapfile.c~mm-swap-move-inode_lock-out-of-claim_swapfile
>> +++ a/mm/swapfile.c
>> @@ -2899,10 +2899,6 @@ static int claim_swapfile(struct swap_in
>> p->bdev = inode->i_sb->s_bdev;
>> }
>>
>> - inode_lock(inode);
>> - if (IS_SWAPFILE(inode))
>> - return -EBUSY;
>> -
>> return 0;
>> }
>>
>> @@ -3157,36 +3153,41 @@ SYSCALL_DEFINE2(swapon, const char __use
>> mapping = swap_file->f_mapping;
>> inode = mapping->host;
>>
>> - /* will take i_rwsem; */
>> error = claim_swapfile(p, inode);
>> if (unlikely(error))
>> goto bad_swap;
>>
>> + inode_lock(inode);
>> + if (IS_SWAPFILE(inode)) {
>> + error = -EBUSY;
>> + goto bad_swap_unlock_inode;
>> + }
>> +
>> /*
>> * Read the swap header.
>> */
>> if (!mapping->a_ops->readpage) {
>> error = -EINVAL;
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>> page = read_mapping_page(mapping, 0, swap_file);
>> if (IS_ERR(page)) {
>> error = PTR_ERR(page);
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>> swap_header = kmap(page);
>>
>> maxpages = read_swap_header(p, swap_header, inode);
>> if (unlikely(!maxpages)) {
>> error = -EINVAL;
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>>
>> /* OK, set up the swap map and apply the bad block list */
>> swap_map = vzalloc(maxpages);
>> if (!swap_map) {
>> error = -ENOMEM;
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>>
>> if (bdi_cap_stable_pages_required(inode_to_bdi(inode)))
>> @@ -3211,7 +3212,7 @@ SYSCALL_DEFINE2(swapon, const char __use
>> GFP_KERNEL);
>> if (!cluster_info) {
>> error = -ENOMEM;
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>>
>> for (ci = 0; ci < nr_cluster; ci++)
>> @@ -3220,7 +3221,7 @@ SYSCALL_DEFINE2(swapon, const char __use
>> p->percpu_cluster = alloc_percpu(struct percpu_cluster);
>> if (!p->percpu_cluster) {
>> error = -ENOMEM;
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>> for_each_possible_cpu(cpu) {
>> struct percpu_cluster *cluster;
>> @@ -3234,13 +3235,13 @@ SYSCALL_DEFINE2(swapon, const char __use
>>
>> error = swap_cgroup_swapon(p->type, maxpages);
>> if (error)
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>>
>> nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map,
>> cluster_info, maxpages, &span);
>> if (unlikely(nr_extents < 0)) {
>> error = nr_extents;
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>> /* frontswap enabled? set up bit-per-page map for frontswap */
>> if (IS_ENABLED(CONFIG_FRONTSWAP))
>> @@ -3280,7 +3281,7 @@ SYSCALL_DEFINE2(swapon, const char __use
>>
>> error = init_swap_address_space(p->type, maxpages);
>> if (error)
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>>
>> /*
>> * Flush any pending IO and dirty mappings before we start using this
>> @@ -3290,7 +3291,7 @@ SYSCALL_DEFINE2(swapon, const char __use
>> error = inode_drain_writes(inode);
>> if (error) {
>> inode->i_flags &= ~S_SWAPFILE;
>> - goto bad_swap;
>> + goto bad_swap_unlock_inode;
>> }
>>
>> mutex_lock(&swapon_mutex);
>> @@ -3315,6 +3316,8 @@ SYSCALL_DEFINE2(swapon, const char __use
>>
>> error = 0;
>> goto out;
>> +bad_swap_unlock_inode:
>> + inode_unlock(inode);
>> bad_swap:
>> free_percpu(p->percpu_cluster);
>> p->percpu_cluster = NULL;
>> @@ -3322,6 +3325,7 @@ bad_swap:
>> set_blocksize(p->bdev, p->old_block_size);
>> blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
>> }
>> + inode = NULL;
>> destroy_swap_extents(p);
>> swap_cgroup_swapoff(p->type);
>> spin_lock(&swap_lock);
>> @@ -3333,13 +3337,8 @@ bad_swap:
>> kvfree(frontswap_map);
>> if (inced_nr_rotate_swap)
>> atomic_dec(&nr_rotate_swap);
>> - if (swap_file) {
>> - if (inode) {
>> - inode_unlock(inode);
>> - inode = NULL;
>> - }
>> + if (swap_file)
>> filp_close(swap_file, NULL);
>> - }
>> out:
>> if (page && !IS_ERR(page)) {
>> kunmap(page);
>> _
>>
On 03/23/20 14:06, Andrew Morton wrote:
> On Mon, 23 Mar 2020 15:30:45 +0000 Qais Yousef <[email protected]> wrote:
>
> > On 03/23/20 15:17, Qais Yousef wrote:
> > > Hi
> > >
> > > I hit the following 2 warnings when running with LOCKDEP=y on arm64 platform
> > > (juno-r2), running on v5.6-rc6
> > >
> > > The 1st one is when I execute `swapon -a`. The 2nd one happens at boot. I have
> > > /dev/sda2 as my swap in /etc/fstab
> > >
> > > Note that I either hit 1 OR 2, but didn't hit both warnings at the same time,
> > > yet at least.
> > >
> > > /dev/sda2 is a usb flash drive, in case it matters somehow.
> >
> > By the way, I noticed that in claim_swapfile() if we fail we don't release the
> > lock. Shouldn't we release the lock here?
> >
> > I tried with that FWIW, but it had no effect on the warnings.
> >
>
> I'll be sending the below into Linus this week.
>
> I was hoping to hear from Darrick/Christoph (?) but it looks like the
> right thing to do. Are you able to test it?
>
> I think I'll add a cc:stable to this one.
>
>
>
> From: Naohiro Aota <[email protected]>
> Subject: mm/swapfile.c: move inode_lock out of claim_swapfile
>
> claim_swapfile() currently keeps the inode locked when it is successful,
> or the file is already swapfile (with -EBUSY). And, on the other error
> cases, it does not lock the inode.
>
> This inconsistency of the lock state and return value is quite confusing
> and actually causing a bad unlock balance as below in the "bad_swap"
> section of __do_sys_swapon().
>
> This commit fixes this issue by moving the inode_lock() and IS_SWAPFILE
> check out of claim_swapfile(). The inode is unlocked in
> "bad_swap_unlock_inode" section, so that the inode is ensured to be
> unlocked at "bad_swap". Thus, error handling codes after the locking now
> jumps to "bad_swap_unlock_inode" instead of "bad_swap".
>
> =====================================
> WARNING: bad unlock balance detected!
> 5.5.0-rc7+ #176 Not tainted
> -------------------------------------
> swapon/4294 is trying to release lock (&sb->s_type->i_mutex_key) at:
> [<ffffffff8173a6eb>] __do_sys_swapon+0x94b/0x3550
> but there are no more locks to release!
>
> other info that might help us debug this:
> no locks held by swapon/4294.
>
> stack backtrace:
> CPU: 5 PID: 4294 Comm: swapon Not tainted 5.5.0-rc7-BTRFS-ZNS+ #176
> Hardware name: ASUS All Series/H87-PRO, BIOS 2102 07/29/2014
> Call Trace:
> dump_stack+0xa1/0xea
> ? __do_sys_swapon+0x94b/0x3550
> print_unlock_imbalance_bug.cold+0x114/0x123
> ? __do_sys_swapon+0x94b/0x3550
> lock_release+0x562/0xed0
> ? kvfree+0x31/0x40
> ? lock_downgrade+0x770/0x770
> ? kvfree+0x31/0x40
> ? rcu_read_lock_sched_held+0xa1/0xd0
> ? rcu_read_lock_bh_held+0xb0/0xb0
> up_write+0x2d/0x490
> ? kfree+0x293/0x2f0
> __do_sys_swapon+0x94b/0x3550
> ? putname+0xb0/0xf0
> ? kmem_cache_free+0x2e7/0x370
> ? do_sys_open+0x184/0x3e0
> ? generic_max_swapfile_size+0x40/0x40
> ? do_syscall_64+0x27/0x4b0
> ? entry_SYSCALL_64_after_hwframe+0x49/0xbe
> ? lockdep_hardirqs_on+0x38c/0x590
> __x64_sys_swapon+0x54/0x80
> do_syscall_64+0xa4/0x4b0
> entry_SYSCALL_64_after_hwframe+0x49/0xbe
> RIP: 0033:0x7f15da0a0dc7
>
> Link: http://lkml.kernel.org/r/[email protected]
> Fixes: 1638045c3677 ("mm: set S_SWAPFILE on blockdev swap devices")
> Signed-off-by: Naohiro Aota <[email protected]>
> Reviewed-by: Andrew Morton <[email protected]>
> Cc: Darrick J. Wong <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> Signed-off-by: Andrew Morton <[email protected]>
> ---
With this patch applied I can no longer reproduce the problem.
Tested-by: Qais Youef <[email protected]>
Thanks!
--
Qais Yousef
>
> mm/swapfile.c | 41 ++++++++++++++++++++---------------------
> 1 file changed, 20 insertions(+), 21 deletions(-)
>
> --- a/mm/swapfile.c~mm-swap-move-inode_lock-out-of-claim_swapfile
> +++ a/mm/swapfile.c
> @@ -2899,10 +2899,6 @@ static int claim_swapfile(struct swap_in
> p->bdev = inode->i_sb->s_bdev;
> }
>
> - inode_lock(inode);
> - if (IS_SWAPFILE(inode))
> - return -EBUSY;
> -
> return 0;
> }
>
> @@ -3157,36 +3153,41 @@ SYSCALL_DEFINE2(swapon, const char __use
> mapping = swap_file->f_mapping;
> inode = mapping->host;
>
> - /* will take i_rwsem; */
> error = claim_swapfile(p, inode);
> if (unlikely(error))
> goto bad_swap;
>
> + inode_lock(inode);
> + if (IS_SWAPFILE(inode)) {
> + error = -EBUSY;
> + goto bad_swap_unlock_inode;
> + }
> +
> /*
> * Read the swap header.
> */
> if (!mapping->a_ops->readpage) {
> error = -EINVAL;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> page = read_mapping_page(mapping, 0, swap_file);
> if (IS_ERR(page)) {
> error = PTR_ERR(page);
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> swap_header = kmap(page);
>
> maxpages = read_swap_header(p, swap_header, inode);
> if (unlikely(!maxpages)) {
> error = -EINVAL;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> /* OK, set up the swap map and apply the bad block list */
> swap_map = vzalloc(maxpages);
> if (!swap_map) {
> error = -ENOMEM;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> if (bdi_cap_stable_pages_required(inode_to_bdi(inode)))
> @@ -3211,7 +3212,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> GFP_KERNEL);
> if (!cluster_info) {
> error = -ENOMEM;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> for (ci = 0; ci < nr_cluster; ci++)
> @@ -3220,7 +3221,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> p->percpu_cluster = alloc_percpu(struct percpu_cluster);
> if (!p->percpu_cluster) {
> error = -ENOMEM;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> for_each_possible_cpu(cpu) {
> struct percpu_cluster *cluster;
> @@ -3234,13 +3235,13 @@ SYSCALL_DEFINE2(swapon, const char __use
>
> error = swap_cgroup_swapon(p->type, maxpages);
> if (error)
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
>
> nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map,
> cluster_info, maxpages, &span);
> if (unlikely(nr_extents < 0)) {
> error = nr_extents;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
> /* frontswap enabled? set up bit-per-page map for frontswap */
> if (IS_ENABLED(CONFIG_FRONTSWAP))
> @@ -3280,7 +3281,7 @@ SYSCALL_DEFINE2(swapon, const char __use
>
> error = init_swap_address_space(p->type, maxpages);
> if (error)
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
>
> /*
> * Flush any pending IO and dirty mappings before we start using this
> @@ -3290,7 +3291,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> error = inode_drain_writes(inode);
> if (error) {
> inode->i_flags &= ~S_SWAPFILE;
> - goto bad_swap;
> + goto bad_swap_unlock_inode;
> }
>
> mutex_lock(&swapon_mutex);
> @@ -3315,6 +3316,8 @@ SYSCALL_DEFINE2(swapon, const char __use
>
> error = 0;
> goto out;
> +bad_swap_unlock_inode:
> + inode_unlock(inode);
> bad_swap:
> free_percpu(p->percpu_cluster);
> p->percpu_cluster = NULL;
> @@ -3322,6 +3325,7 @@ bad_swap:
> set_blocksize(p->bdev, p->old_block_size);
> blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
> }
> + inode = NULL;
> destroy_swap_extents(p);
> swap_cgroup_swapoff(p->type);
> spin_lock(&swap_lock);
> @@ -3333,13 +3337,8 @@ bad_swap:
> kvfree(frontswap_map);
> if (inced_nr_rotate_swap)
> atomic_dec(&nr_rotate_swap);
> - if (swap_file) {
> - if (inode) {
> - inode_unlock(inode);
> - inode = NULL;
> - }
> + if (swap_file)
> filp_close(swap_file, NULL);
> - }
> out:
> if (page && !IS_ERR(page)) {
> kunmap(page);
> _
>
On Tue, Mar 24, 2020 at 06:59:08AM +0900, Naohiro Aota wrote:
> On Mon, Mar 23, 2020 at 02:12:01PM -0700, Darrick J. Wong wrote:
> > On Mon, Mar 23, 2020 at 02:06:15PM -0700, Andrew Morton wrote:
> > > On Mon, 23 Mar 2020 15:30:45 +0000 Qais Yousef <[email protected]> wrote:
> > >
> > > > On 03/23/20 15:17, Qais Yousef wrote:
> > > > > Hi
> > > > >
> > > > > I hit the following 2 warnings when running with LOCKDEP=y on arm64 platform
> > > > > (juno-r2), running on v5.6-rc6
> > > > >
> > > > > The 1st one is when I execute `swapon -a`. The 2nd one happens at boot. I have
> > > > > /dev/sda2 as my swap in /etc/fstab
> > > > >
> > > > > Note that I either hit 1 OR 2, but didn't hit both warnings at the same time,
> > > > > yet at least.
> > > > >
> > > > > /dev/sda2 is a usb flash drive, in case it matters somehow.
> > > >
> > > > By the way, I noticed that in claim_swapfile() if we fail we don't release the
> > > > lock. Shouldn't we release the lock here?
> > > >
> > > > I tried with that FWIW, but it had no effect on the warnings.
> > > >
> > >
> > > I'll be sending the below into Linus this week.
> > >
> > > I was hoping to hear from Darrick/Christoph (?) but it looks like the
> > > right thing to do. Are you able to test it?
> >
> > I had questions[1] about the patch, but nobody ever replied.
>
> Sorry, I have overlooked that one. I'm replying here.
>
> > Sorry to wander in late, but I don't see how we unlock the inode in the
> > success case. Before this patch, the "if (inode) inode_unlock(inode);"
> > below out: would take care of this for both the success case and the
> > bad_swap case, but now that's gone, and AFAICT after this patch we only
> > unlock the inode when erroring out...
> >
> > > +bad_swap_unlock_inode:
> > > + inode_unlock(inode);
> >
> > ...since we never goto bad_swap_unlock_inode when error == 0, correct?
>
> Actually, this patch does not touch "if (inode) inode_unlock(inode);" below
> the "out:" label, so we still have that "inode_unlock" after this patch and
> it unlocks the inode in the success case.
Ah, right, I don't know what I was thinking when I wrote that. :)
Reviewed-by: Darrick J. Wong <[email protected]>
--D
>
> >
> > --D
> >
> > [1] https://lore.kernel.org/linux-fsdevel/20200212164956.GK6874@magnolia/
> >
> > > I think I'll add a cc:stable to this one.
> > >
> > >
> > >
> > > From: Naohiro Aota <[email protected]>
> > > Subject: mm/swapfile.c: move inode_lock out of claim_swapfile
> > >
> > > claim_swapfile() currently keeps the inode locked when it is successful,
> > > or the file is already swapfile (with -EBUSY). And, on the other error
> > > cases, it does not lock the inode.
> > >
> > > This inconsistency of the lock state and return value is quite confusing
> > > and actually causing a bad unlock balance as below in the "bad_swap"
> > > section of __do_sys_swapon().
> > >
> > > This commit fixes this issue by moving the inode_lock() and IS_SWAPFILE
> > > check out of claim_swapfile(). The inode is unlocked in
> > > "bad_swap_unlock_inode" section, so that the inode is ensured to be
> > > unlocked at "bad_swap". Thus, error handling codes after the locking now
> > > jumps to "bad_swap_unlock_inode" instead of "bad_swap".
> > >
> > > =====================================
> > > WARNING: bad unlock balance detected!
> > > 5.5.0-rc7+ #176 Not tainted
> > > -------------------------------------
> > > swapon/4294 is trying to release lock (&sb->s_type->i_mutex_key) at:
> > > [<ffffffff8173a6eb>] __do_sys_swapon+0x94b/0x3550
> > > but there are no more locks to release!
> > >
> > > other info that might help us debug this:
> > > no locks held by swapon/4294.
> > >
> > > stack backtrace:
> > > CPU: 5 PID: 4294 Comm: swapon Not tainted 5.5.0-rc7-BTRFS-ZNS+ #176
> > > Hardware name: ASUS All Series/H87-PRO, BIOS 2102 07/29/2014
> > > Call Trace:
> > > dump_stack+0xa1/0xea
> > > ? __do_sys_swapon+0x94b/0x3550
> > > print_unlock_imbalance_bug.cold+0x114/0x123
> > > ? __do_sys_swapon+0x94b/0x3550
> > > lock_release+0x562/0xed0
> > > ? kvfree+0x31/0x40
> > > ? lock_downgrade+0x770/0x770
> > > ? kvfree+0x31/0x40
> > > ? rcu_read_lock_sched_held+0xa1/0xd0
> > > ? rcu_read_lock_bh_held+0xb0/0xb0
> > > up_write+0x2d/0x490
> > > ? kfree+0x293/0x2f0
> > > __do_sys_swapon+0x94b/0x3550
> > > ? putname+0xb0/0xf0
> > > ? kmem_cache_free+0x2e7/0x370
> > > ? do_sys_open+0x184/0x3e0
> > > ? generic_max_swapfile_size+0x40/0x40
> > > ? do_syscall_64+0x27/0x4b0
> > > ? entry_SYSCALL_64_after_hwframe+0x49/0xbe
> > > ? lockdep_hardirqs_on+0x38c/0x590
> > > __x64_sys_swapon+0x54/0x80
> > > do_syscall_64+0xa4/0x4b0
> > > entry_SYSCALL_64_after_hwframe+0x49/0xbe
> > > RIP: 0033:0x7f15da0a0dc7
> > >
> > > Link: http://lkml.kernel.org/r/[email protected]
> > > Fixes: 1638045c3677 ("mm: set S_SWAPFILE on blockdev swap devices")
> > > Signed-off-by: Naohiro Aota <[email protected]>
> > > Reviewed-by: Andrew Morton <[email protected]>
> > > Cc: Darrick J. Wong <[email protected]>
> > > Cc: Christoph Hellwig <[email protected]>
> > > Signed-off-by: Andrew Morton <[email protected]>
> > > ---
> > >
> > > mm/swapfile.c | 41 ++++++++++++++++++++---------------------
> > > 1 file changed, 20 insertions(+), 21 deletions(-)
> > >
> > > --- a/mm/swapfile.c~mm-swap-move-inode_lock-out-of-claim_swapfile
> > > +++ a/mm/swapfile.c
> > > @@ -2899,10 +2899,6 @@ static int claim_swapfile(struct swap_in
> > > p->bdev = inode->i_sb->s_bdev;
> > > }
> > >
> > > - inode_lock(inode);
> > > - if (IS_SWAPFILE(inode))
> > > - return -EBUSY;
> > > -
> > > return 0;
> > > }
> > >
> > > @@ -3157,36 +3153,41 @@ SYSCALL_DEFINE2(swapon, const char __use
> > > mapping = swap_file->f_mapping;
> > > inode = mapping->host;
> > >
> > > - /* will take i_rwsem; */
> > > error = claim_swapfile(p, inode);
> > > if (unlikely(error))
> > > goto bad_swap;
> > >
> > > + inode_lock(inode);
> > > + if (IS_SWAPFILE(inode)) {
> > > + error = -EBUSY;
> > > + goto bad_swap_unlock_inode;
> > > + }
> > > +
> > > /*
> > > * Read the swap header.
> > > */
> > > if (!mapping->a_ops->readpage) {
> > > error = -EINVAL;
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > > page = read_mapping_page(mapping, 0, swap_file);
> > > if (IS_ERR(page)) {
> > > error = PTR_ERR(page);
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > > swap_header = kmap(page);
> > >
> > > maxpages = read_swap_header(p, swap_header, inode);
> > > if (unlikely(!maxpages)) {
> > > error = -EINVAL;
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > >
> > > /* OK, set up the swap map and apply the bad block list */
> > > swap_map = vzalloc(maxpages);
> > > if (!swap_map) {
> > > error = -ENOMEM;
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > >
> > > if (bdi_cap_stable_pages_required(inode_to_bdi(inode)))
> > > @@ -3211,7 +3212,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> > > GFP_KERNEL);
> > > if (!cluster_info) {
> > > error = -ENOMEM;
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > >
> > > for (ci = 0; ci < nr_cluster; ci++)
> > > @@ -3220,7 +3221,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> > > p->percpu_cluster = alloc_percpu(struct percpu_cluster);
> > > if (!p->percpu_cluster) {
> > > error = -ENOMEM;
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > > for_each_possible_cpu(cpu) {
> > > struct percpu_cluster *cluster;
> > > @@ -3234,13 +3235,13 @@ SYSCALL_DEFINE2(swapon, const char __use
> > >
> > > error = swap_cgroup_swapon(p->type, maxpages);
> > > if (error)
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > >
> > > nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map,
> > > cluster_info, maxpages, &span);
> > > if (unlikely(nr_extents < 0)) {
> > > error = nr_extents;
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > > /* frontswap enabled? set up bit-per-page map for frontswap */
> > > if (IS_ENABLED(CONFIG_FRONTSWAP))
> > > @@ -3280,7 +3281,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> > >
> > > error = init_swap_address_space(p->type, maxpages);
> > > if (error)
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > >
> > > /*
> > > * Flush any pending IO and dirty mappings before we start using this
> > > @@ -3290,7 +3291,7 @@ SYSCALL_DEFINE2(swapon, const char __use
> > > error = inode_drain_writes(inode);
> > > if (error) {
> > > inode->i_flags &= ~S_SWAPFILE;
> > > - goto bad_swap;
> > > + goto bad_swap_unlock_inode;
> > > }
> > >
> > > mutex_lock(&swapon_mutex);
> > > @@ -3315,6 +3316,8 @@ SYSCALL_DEFINE2(swapon, const char __use
> > >
> > > error = 0;
> > > goto out;
> > > +bad_swap_unlock_inode:
> > > + inode_unlock(inode);
> > > bad_swap:
> > > free_percpu(p->percpu_cluster);
> > > p->percpu_cluster = NULL;
> > > @@ -3322,6 +3325,7 @@ bad_swap:
> > > set_blocksize(p->bdev, p->old_block_size);
> > > blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL);
> > > }
> > > + inode = NULL;
> > > destroy_swap_extents(p);
> > > swap_cgroup_swapoff(p->type);
> > > spin_lock(&swap_lock);
> > > @@ -3333,13 +3337,8 @@ bad_swap:
> > > kvfree(frontswap_map);
> > > if (inced_nr_rotate_swap)
> > > atomic_dec(&nr_rotate_swap);
> > > - if (swap_file) {
> > > - if (inode) {
> > > - inode_unlock(inode);
> > > - inode = NULL;
> > > - }
> > > + if (swap_file)
> > > filp_close(swap_file, NULL);
> > > - }
> > > out:
> > > if (page && !IS_ERR(page)) {
> > > kunmap(page);
> > > _
> > >