Re: [PATCH] NFS: simplify check for freeing cn_resp

[Trond Myklebust <trondmy@hammerspace.com>]

Hi Tom,

On Sun, 2022-01-16 at 06:43 -0800, trix@redhat.com wrote:
> From: Tom Rix <trix@redhat.com>
> 
> nfs42_files_from_same_server() is called to check if freeing
> cn_resp is required.  Instead of calling a function, check
> the pointer.
> 
> Signed-off-by: Tom Rix <trix@redhat.com>
> ---
>  fs/nfs/nfs4file.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c
> index e79ae4cbc395e..677631ea4cfb3 100644
> --- a/fs/nfs/nfs4file.c
> +++ b/fs/nfs/nfs4file.c
> @@ -180,7 +180,7 @@ static ssize_t __nfs4_copy_file_range(struct file
> *file_in, loff_t pos_in,
>         ret = nfs42_proc_copy(file_in, pos_in, file_out, pos_out,
> count,
>                                 nss, cnrs, sync);
>  out:
> -       if (!nfs42_files_from_same_server(file_in, file_out))
> +       if (cn_resp)
>                 kfree(cn_resp);

The kernel convention in these circumstances is to just skip the NULL
pointer check, since kfree() does that anyway.

>         if (ret == -EAGAIN)
>                 goto retry;

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