Re: [PATCH v2] mm, swap: move inode_lock out of claim_swapfile

From: Andrew Morton <akpm@linux-foundation.org>
To: Naohiro Aota <naohiro.aota@wdc.com>
Cc: linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
	linux-block@vger.kernel.org,
	Christoph Hellwig <hch@infradead.org>,
	"Darrick J . Wong" <darrick.wong@oracle.com>
Subject: Re: [PATCH v2] mm, swap: move inode_lock out of claim_swapfile
Date: Sun, 9 Feb 2020 20:16:12 -0800	[thread overview]
Message-ID: <20200209201612.e5f234b357823df574104cb9@linux-foundation.org> (raw)
In-Reply-To: <20200206090132.154869-1-naohiro.aota@wdc.com>

On Thu,  6 Feb 2020 18:01:32 +0900 Naohiro Aota <naohiro.aota@wdc.com> wrote:

> 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
> 
>  mm/swapfile.c | 41 ++++++++++++++++++++---------------------
>  1 file changed, 20 insertions(+), 21 deletions(-)
> 
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index bb3261d45b6a..2c4c349e1101 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c

Look correct to me.

But I don't think this code at the end of sys_swapon():

	if (inode)
		inode_unlock(inode);

will ever execute?  `inode' is always NULL here?