Re: [PATCH 2/3] pNFS: honor clora_changed when recalling a layout

Hi Ben,

On Wed, Jun 24, 2026, at 3:50 PM, Benjamin Coddington wrote:
> When the metadata server recalls a layout with clora_changed FALSE, the
> layout is not changing and the client may complete its modified writes to
> the storage devices before returning the layout (RFC 8881, Section
> 20.3.3).  Only when clora_changed is TRUE -- the server is restriping, or
> a storage device has failed -- should the client stop writing to the
> storage devices and redirect through the metadata server.
>
> Since commit b739a5bd9d9f ("NFSv4/flexfiles: Cancel I/O if the layout is
> recalled or revoked") the client cancels in-flight I/O on every recall,
> regardless of clora_changed.  For an unchanged recall this abandons
> writes whose data may already have reached the storage device; such a
> write can then land after the LAYOUTRETURN, which the server sees as a
> write without a layout.
>
> Pass the recall's clora_changed value through
> pnfs_mark_matching_lsegs_return() and only cancel in-flight I/O when the
> layout is actually changing.  When it is not, the existing deferred
> return path waits for the in-flight writes to drain before sending the
> LAYOUTRETURN.  Other callers, which are tearing down or returning the
> layout for their own reasons, continue to cancel as before.
>
> Signed-off-by: Benjamin Coddington <bcodding-F/[email protected]>
> ---
>  fs/nfs/callback_proc.c |  3 ++-
>  fs/nfs/pnfs.c          | 21 ++++++++++++---------
>  fs/nfs/pnfs.h          |  2 +-
>  3 files changed, 15 insertions(+), 11 deletions(-)
>
> diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
> index 021572312b0e..7725daed2697 100644
> --- a/fs/nfs/callback_proc.c
> +++ b/fs/nfs/callback_proc.c
> @@ -290,7 +290,8 @@ static u32 initiate_file_draining(struct nfs_client 
> *clp,
>  	pnfs_set_layout_stateid(lo, &args->cbl_stateid, NULL, true);
>  	switch (pnfs_mark_matching_lsegs_return(lo, &free_me_list,
>  				&args->cbl_range,
> -				be32_to_cpu(args->cbl_stateid.seqid))) {
> +				be32_to_cpu(args->cbl_stateid.seqid),
> +				args->cbl_layoutchanged)) {
>  	case 0:
>  	case -EBUSY:
>  		/* There are layout segments that need to be returned */
> diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
> index 743467e9ba20..614e1b28bec0 100644
> --- a/fs/nfs/pnfs.c
> +++ b/fs/nfs/pnfs.c
> @@ -432,7 +432,8 @@ bool nfs4_layout_refresh_old_stateid(nfs4_stateid 
> *dst,
>  			goto out;
>  		}
>  		/* Try to update the seqid to the most recent */
> -		err = pnfs_mark_matching_lsegs_return(lo, &head, &range, 0);
> +		err = pnfs_mark_matching_lsegs_return(lo, &head, &range, 0,
> +						      true);
>  		if (err != -EBUSY) {
>  			dst->seqid = lo->plh_stateid.seqid;
>  			*dst_range = range;
> @@ -486,7 +487,7 @@ static int pnfs_mark_layout_stateid_return(struct 
> pnfs_layout_hdr *lo,
>  		.length = NFS4_MAX_UINT64,
>  	};
> 
> -	return pnfs_mark_matching_lsegs_return(lo, lseg_list, &range, seq);
> +	return pnfs_mark_matching_lsegs_return(lo, lseg_list, &range, seq, true);
>  }
> 
>  static int
> @@ -524,7 +525,7 @@ pnfs_layout_io_set_failed(struct pnfs_layout_hdr 
> *lo, u32 iomode)
> 
>  	spin_lock(&inode->i_lock);
>  	pnfs_layout_set_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode));
> -	pnfs_mark_matching_lsegs_return(lo, &head, &range, 0);
> +	pnfs_mark_matching_lsegs_return(lo, &head, &range, 0, true);
>  	spin_unlock(&inode->i_lock);
>  	pnfs_free_lseg_list(&head);
>  	dprintk("%s Setting layout IOMODE_%s fail bit\n", __func__,
> @@ -1461,7 +1462,7 @@ _pnfs_return_layout(struct inode *ino)
>  	}
>  	valid_layout = pnfs_layout_is_valid(lo);
>  	pnfs_clear_layoutcommit(ino, &tmp_list);
> -	pnfs_mark_matching_lsegs_return(lo, &tmp_list, &range, 0);
> +	pnfs_mark_matching_lsegs_return(lo, &tmp_list, &range, 0, true);
> 
>  	if (NFS_SERVER(ino)->pnfs_curr_ld->return_range)
>  		NFS_SERVER(ino)->pnfs_curr_ld->return_range(lo, &range);
> @@ -2621,7 +2622,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
>  			.iomode = IOMODE_ANY,
>  			.length = NFS4_MAX_UINT64,
>  		};
> -		pnfs_mark_matching_lsegs_return(lo, &free_me, &range, 0);
> +		pnfs_mark_matching_lsegs_return(lo, &free_me, &range, 0, true);
>  		goto out_forget;
>  	} else {
>  		/* We have a completely new layout */
> @@ -2666,7 +2667,7 @@ int
>  pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
>  				struct list_head *tmp_list,
>  				const struct pnfs_layout_range *return_range,
> -				u32 seq)
> +				u32 seq, bool cancel_io)

Can you also add a description for the new "cancel_io" parameter to the
documentation right above this function?

Thanks,
Anna

>  {
>  	struct pnfs_layout_segment *lseg, *next;
>  	struct nfs_server *server = NFS_SERVER(lo->plh_inode);
> @@ -2692,7 +2693,8 @@ pnfs_mark_matching_lsegs_return(struct 
> pnfs_layout_hdr *lo,
>  				continue;
>  			remaining++;
>  			set_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags);
> -			pnfs_lseg_cancel_io(server, lseg);
> +			if (cancel_io)
> +				pnfs_lseg_cancel_io(server, lseg);
>  		}
> 
>  	if (remaining) {
> @@ -2727,7 +2729,8 @@ pnfs_mark_layout_for_return(struct inode *inode,
>  	 * segments at hand when sending layoutreturn. See pnfs_put_lseg()
>  	 * for how it works.
>  	 */
> -	if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs, range, 
> 0) != -EBUSY) {
> +	if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs, range, 
> 0,
> +					    true) != -EBUSY) {
>  		const struct cred *cred;
>  		nfs4_stateid stateid;
>  		enum pnfs_iomode iomode;
> @@ -2842,7 +2845,7 @@ static int 
> pnfs_layout_return_unused_byserver(struct nfs_server *server,
>  		pnfs_get_layout_hdr(lo);
>  		pnfs_set_plh_return_info(lo, range->iomode, 0);
>  		if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs,
> -						    range, 0) != 0 ||
> +						    range, 0, true) != 0 ||
>  		    !pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode)) {
>  			spin_unlock(&inode->i_lock);
>  			rcu_read_unlock();
> diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
> index eb39859c216c..673c2b244978 100644
> --- a/fs/nfs/pnfs.h
> +++ b/fs/nfs/pnfs.h
> @@ -300,7 +300,7 @@ int pnfs_mark_matching_lsegs_invalid(struct 
> pnfs_layout_hdr *lo,
>  int pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
>  				struct list_head *tmp_list,
>  				const struct pnfs_layout_range *recall_range,
> -				u32 seq);
> +				u32 seq, bool cancel_io);
>  int pnfs_mark_layout_stateid_invalid(struct pnfs_layout_hdr *lo,
>  		struct list_head *lseg_list);
>  bool pnfs_roc(struct inode *ino, struct nfs4_layoutreturn_args *args,
> -- 
> 2.53.0