DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 89B7E20869
Message-ID: <4e5c759108926983cdbbdedf24995db84779ef2e.camel@kernel.org>
Subject: Re: [PATCH v1 03/11] NFS test for intra vs inter COPY
From:   Jeff Layton <jlayton@kernel.org>
To:     Olga Kornievskaia <olga.kornievskaia@gmail.com>,
        trondmy@hammerspace.com, anna.schumaker@netapp.com
Cc:     linux-nfs@vger.kernel.org
Date:   Sun, 21 Oct 2018 10:44:14 -0400
In-Reply-To: <20181019152932.32462-4-olga.kornievskaia@gmail.com>
References: <20181019152932.32462-1-olga.kornievskaia@gmail.com>
         <20181019152932.32462-4-olga.kornievskaia@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-nfs-owner@vger.kernel.org
Precedence: bulk

On Fri, 2018-10-19 at 11:29 -0400, Olga Kornievskaia wrote:
> From: Olga Kornievskaia <kolga@netapp.com>
> 
> Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
> ---
>  fs/nfs/nfs42.h | 11 +++++++++++
>  1 file changed, 11 insertions(+)
> 
> diff --git a/fs/nfs/nfs42.h b/fs/nfs/nfs42.h
> index 19ec38f8..5abff4d 100644
> --- a/fs/nfs/nfs42.h
> +++ b/fs/nfs/nfs42.h
> @@ -6,6 +6,7 @@
>  #ifndef __LINUX_FS_NFS_NFS4_2_H
>  #define __LINUX_FS_NFS_NFS4_2_H
>  
> +#include <linux/sunrpc/addr.h>
>  /*
>   * FIXME:  four LAYOUTSTATS calls per compound at most! Do we need to support
>   * more? Need to consider not to pre-alloc too much for a compound.
> @@ -21,4 +22,14 @@ int nfs42_proc_layoutstats_generic(struct nfs_server *,
>  				   struct nfs42_layoutstat_data *);
>  int nfs42_proc_clone(struct file *, struct file *, loff_t, loff_t, loff_t);
>  
> +static inline bool nfs42_files_from_same_server(struct file *in,
> +						struct file *out)
> +{
> +	struct nfs_client *c_in = (NFS_SERVER(file_inode(in)))->nfs_client;
> +	struct nfs_client *c_out = (NFS_SERVER(file_inode(out)))->nfs_client;
> +
> +	return rpc_cmp_addr((struct sockaddr *)&c_in->cl_addr,
> +				(struct sockaddr *)&c_out->cl_addr);
> +}
> +
>  #endif /* __LINUX_FS_NFS_NFS4_2_H */

What about trunking? If the addresses don't match, should you compare
cl_serverowner or something too? Or maybe just do that instead of
testing addresses?

It's usually best to add infrastructure like this in the same patch with
an initial caller so we can see how it's intended to be used. 
-- 
Jeff Layton <jlayton@kernel.org>