Received-SPF: pass (google.com: domain of linux-kernel+bounces-66965-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Thu, 15 Feb 2024 14:16:38 +0100
From: Jan Kara <jack@suse.cz>
To: Chuck Lever <cel@kernel.org>
Cc: viro@zeniv.linux.org.uk, brauner@kernel.org, jack@suse.cz,
	hughd@google.com, akpm@linux-foundation.org,
	Liam.Howlett@oracle.com, oliver.sang@intel.com, feng.tang@intel.com,
	linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	maple-tree@lists.infradead.org, linux-mm@kvack.org, lkp@intel.com
Subject: Re: [PATCH RFC 7/7] libfs: Re-arrange locking in offset_iterate_dir()
Message-ID: <20240215131638.cxipaxanhidb3pev@quack3>
References: <170785993027.11135.8830043889278631735.stgit@91.116.238.104.host.secureserver.net>
 <170786028847.11135.14775608389430603086.stgit@91.116.238.104.host.secureserver.net>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <170786028847.11135.14775608389430603086.stgit@91.116.238.104.host.secureserver.net>

On Tue 13-02-24 16:38:08, Chuck Lever wrote:
> From: Chuck Lever <chuck.lever@oracle.com>
> 
> Liam says that, unlike with xarray, once the RCU read lock is
> released ma_state is not safe to re-use for the next mas_find() call.
> But the RCU read lock has to be released on each loop iteration so
> that dput() can be called safely.
> 
> Thus we are forced to walk the offset tree with fresh state for each
> directory entry. mt_find() can do this for us, though it might be a
> little less efficient than maintaining ma_state locally.
> 
> Since offset_iterate_dir() doesn't build ma_state locally any more,
> there's no longer a strong need for offset_find_next(). Clean up by
> rolling these two helpers together.
> 
> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>

Well, in general I think even xas_next_entry() is not safe to use how
offset_find_next() was using it. Once you drop rcu_read_lock(),
xas->xa_node could go stale. But since you're holding inode->i_rwsem when
using offset_find_next() you should be protected from concurrent
modifications of the mapping (whatever the underlying data structure is) -
that's what makes xas_next_entry() safe AFAIU. Isn't that enough for the
maple tree? Am I missing something?

								Honza

> ---
>  fs/libfs.c |   39 +++++++++++++++++----------------------
>  1 file changed, 17 insertions(+), 22 deletions(-)
> 
> diff --git a/fs/libfs.c b/fs/libfs.c
> index f073e9aeb2bf..6e01fde1cf95 100644
> --- a/fs/libfs.c
> +++ b/fs/libfs.c
> @@ -436,23 +436,6 @@ static loff_t offset_dir_llseek(struct file *file, loff_t offset, int whence)
>  	return vfs_setpos(file, offset, MAX_LFS_FILESIZE);
>  }
>  
> -static struct dentry *offset_find_next(struct ma_state *mas)
> -{
> -	struct dentry *child, *found = NULL;
> -
> -	rcu_read_lock();
> -	child = mas_find(mas, ULONG_MAX);
> -	if (!child)
> -		goto out;
> -	spin_lock(&child->d_lock);
> -	if (simple_positive(child))
> -		found = dget_dlock(child);
> -	spin_unlock(&child->d_lock);
> -out:
> -	rcu_read_unlock();
> -	return found;
> -}
> -
>  static bool offset_dir_emit(struct dir_context *ctx, struct dentry *dentry)
>  {
>  	unsigned long offset = dentry2offset(dentry);
> @@ -465,13 +448,22 @@ static bool offset_dir_emit(struct dir_context *ctx, struct dentry *dentry)
>  static void *offset_iterate_dir(struct inode *inode, struct dir_context *ctx)
>  {
>  	struct offset_ctx *octx = inode->i_op->get_offset_ctx(inode);
> -	MA_STATE(mas, &octx->mt, ctx->pos, ctx->pos);
> -	struct dentry *dentry;
> +	struct dentry *dentry, *found;
> +	unsigned long offset;
>  
> +	offset = ctx->pos;
>  	while (true) {
> -		dentry = offset_find_next(&mas);
> +		found = mt_find(&octx->mt, &offset, ULONG_MAX);
> +		if (!found)
> +			goto out_noent;
> +
> +		dentry = NULL;
> +		spin_lock(&found->d_lock);
> +		if (simple_positive(found))
> +			dentry = dget_dlock(found);
> +		spin_unlock(&found->d_lock);
>  		if (!dentry)
> -			return ERR_PTR(-ENOENT);
> +			goto out_noent;
>  
>  		if (!offset_dir_emit(ctx, dentry)) {
>  			dput(dentry);
> @@ -479,9 +471,12 @@ static void *offset_iterate_dir(struct inode *inode, struct dir_context *ctx)
>  		}
>  
>  		dput(dentry);
> -		ctx->pos = mas.index + 1;
> +		ctx->pos = offset;
>  	}
>  	return NULL;
> +
> +out_noent:
> +	return ERR_PTR(-ENOENT);
>  }
>  
>  /**
> 
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR