Re: [PATCH 12/50] fs: rework iput logic

[Christian Brauner <brauner@kernel.org>]

On Thu, Aug 21, 2025 at 04:18:23PM -0400, Josef Bacik wrote:
> Currently, if we are the last iput, and we have the I_DIRTY_TIME bit
> set, we will grab a reference on the inode again and then mark it dirty
> and then redo the put.  This is to make sure we delay the time update
> for as long as possible.
> 
> We can rework this logic to simply dec i_count if it is not 1, and if it
> is do the time update while still holding the i_count reference.
> 
> Then we can replace the atomic_dec_and_lock with locking the ->i_lock
> and doing atomic_dec_and_test, since we did the atomic_add_unless above.
> 
> This is preparation for no longer allowing 0 i_count inodes to exist.
> 
> Signed-off-by: Josef Bacik <josef@toxicpanda.com>
> ---
>  fs/inode.c | 31 ++++++++++++++++---------------
>  1 file changed, 16 insertions(+), 15 deletions(-)
> 
> diff --git a/fs/inode.c b/fs/inode.c
> index 16acad5583fc..814c03f5dbb1 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -1928,22 +1928,23 @@ void iput(struct inode *inode)
>  	if (!inode)
>  		return;
>  	BUG_ON(inode->i_state & I_CLEAR);
> -retry:
> -	if (atomic_dec_and_lock(&inode->i_count, &inode->i_lock)) {
> -		if (inode->i_nlink && (inode->i_state & I_DIRTY_TIME)) {
> -			/*
> -			 * Increment i_count directly as we still have our
> -			 * i_obj_count reference still. This is temporary and
> -			 * will go away in a future patch.
> -			 */
> -			atomic_inc(&inode->i_count);
> -			spin_unlock(&inode->i_lock);
> -			trace_writeback_lazytime_iput(inode);
> -			mark_inode_dirty_sync(inode);
> -			goto retry;
> -		}
> -		iput_final(inode);
> +
> +	if (atomic_add_unless(&inode->i_count, -1, 1)) {
> +		iobj_put(inode);
> +		return;
>  	}
> +
> +	if (inode->i_nlink && (inode->i_state & I_DIRTY_TIME)) {
> +		trace_writeback_lazytime_iput(inode);
> +		mark_inode_dirty_sync(inode);
> +	}
> +
> +	spin_lock(&inode->i_lock);
> +	if (atomic_dec_and_test(&inode->i_count))
> +		iput_final(inode);

Personally, I'd add a
// drops i_lock
comment behind iput_final() but that's a matter of taste tbf.

> +	else
> +		spin_unlock(&inode->i_lock);
> +
>  	iobj_put(inode);
>  }
>  EXPORT_SYMBOL(iput);

This looks a lot less magical than the current variant! We should maybe
split this patch in two. A cleanup patch that removes the questionable
"drop to zero, take lock then increment from zero again" logic and then
in a separate patch add in the iobj_put(). So the cleanup can go to the
front of the series.