Re: [PATCH] nfsd: Prevent truncation of an unlinked inode from blocking access to its directory

[Trond Myklebust <trondmy@hammerspace.com>]

On Fri, 2021-05-14 at 11:58 +0800, Nick Huang wrote:
> From: Yu Hsiang Huang <nickhuang@synology.com>
> 
> Truncation of an unlinked inode may take a long time for I/O waiting,
> and
> it doesn't have to prevent access to the directory. Thus, let
> truncation
> occur outside the directory's mutex, just like do_unlinkat() does.
> 
> Signed-off-by: Yu Hsiang Huang <nickhuang@synology.com>
> Signed-off-by: Bing Jing Chang <bingjingc@synology.com>
> Signed-off-by: Robbie Ko <robbieko@synology.com>
> ---
>  fs/nfsd/vfs.c | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
> index 15adf1f6ab21..39948f130712 100644
> --- a/fs/nfsd/vfs.c
> +++ b/fs/nfsd/vfs.c
> @@ -1859,6 +1859,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct
> svc_fh *fhp, int type,
>  {
>         struct dentry   *dentry, *rdentry;
>         struct inode    *dirp;
> +       struct inode    *rinode;
>         __be32          err;
>         int             host_err;
>  
> @@ -1887,6 +1888,8 @@ nfsd_unlink(struct svc_rqst *rqstp, struct
> svc_fh *fhp, int type,
>                 host_err = -ENOENT;
>                 goto out_drop_write;
>         }
> +       rinode = d_inode(rdentry);
> +       ihold(rinode);
>  
>         if (!type)
>                 type = d_inode(rdentry)->i_mode & S_IFMT;
> @@ -1902,6 +1905,8 @@ nfsd_unlink(struct svc_rqst *rqstp, struct
> svc_fh *fhp, int type,
>         if (!host_err)
>                 host_err = commit_metadata(fhp);

Why leave the commit_metadata() call under the lock? If you're
concerned about latency, then it makes more sense to call fh_unlock()
before flushing those metadata updates to disk.

This is, BTW, an optimisation that appears to be possible in several
other cases in fs/nfsd/vfs.c.

>         dput(rdentry);
> +       fh_unlock(fhp);
> +       iput(rinode);    /* truncate the inode here */
>  
>  out_drop_write:
>         fh_drop_write(fhp);

-- 
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trond.myklebust@hammerspace.com