NFSD: handle GETATTR conflict with write delegation
authorDai Ngo <dai.ngo@oracle.com>
Thu, 15 Feb 2024 22:05:22 +0000 (14:05 -0800)
committerChuck Lever <chuck.lever@oracle.com>
Fri, 1 Mar 2024 14:12:32 +0000 (09:12 -0500)
If the GETATTR request on a file that has write delegation in effect
and the request attributes include the change info and size attribute
then the request is handled as below:

Server sends CB_GETATTR to client to get the latest change info and file
size. If these values are the same as the server's cached values then
the GETATTR proceeds as normal.

If either the change info or file size is different from the server's
cached values, or the file was already marked as modified, then:

    . update time_modify and time_metadata into file's metadata
      with current time

    . encode GETATTR as normal except the file size is encoded with
      the value returned from CB_GETATTR

    . mark the file as modified

If the CB_GETATTR fails for any reasons, the delegation is recalled
and NFS4ERR_DELAY is returned for the GETATTR.

Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
fs/nfsd/nfs4state.c
fs/nfsd/nfs4xdr.c
fs/nfsd/nfsd.h
fs/nfsd/state.h

index 994dcf8e54205d096200b77a4ee354db449ca9b9..a22016d0106fbfe8269a6de5695e7e03cba472bc 100644 (file)
@@ -127,6 +127,7 @@ static void free_session(struct nfsd4_session *);
 
 static const struct nfsd4_callback_ops nfsd4_cb_recall_ops;
 static const struct nfsd4_callback_ops nfsd4_cb_notify_lock_ops;
+static const struct nfsd4_callback_ops nfsd4_cb_getattr_ops;
 
 static struct workqueue_struct *laundry_wq;
 
@@ -1189,6 +1190,10 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_file *fp,
        dp->dl_recalled = false;
        nfsd4_init_cb(&dp->dl_recall, dp->dl_stid.sc_client,
                      &nfsd4_cb_recall_ops, NFSPROC4_CLNT_CB_RECALL);
+       nfsd4_init_cb(&dp->dl_cb_fattr.ncf_getattr, dp->dl_stid.sc_client,
+                       &nfsd4_cb_getattr_ops, NFSPROC4_CLNT_CB_GETATTR);
+       dp->dl_cb_fattr.ncf_file_modified = false;
+       dp->dl_cb_fattr.ncf_cb_bmap[0] = FATTR4_WORD0_CHANGE | FATTR4_WORD0_SIZE;
        get_nfs4_file(fp);
        dp->dl_stid.sc_file = fp;
        return dp;
@@ -3044,11 +3049,59 @@ nfsd4_cb_recall_any_release(struct nfsd4_callback *cb)
        spin_unlock(&nn->client_lock);
 }
 
+static int
+nfsd4_cb_getattr_done(struct nfsd4_callback *cb, struct rpc_task *task)
+{
+       struct nfs4_cb_fattr *ncf =
+                       container_of(cb, struct nfs4_cb_fattr, ncf_getattr);
+
+       ncf->ncf_cb_status = task->tk_status;
+       switch (task->tk_status) {
+       case -NFS4ERR_DELAY:
+               rpc_delay(task, 2 * HZ);
+               return 0;
+       default:
+               return 1;
+       }
+}
+
+static void
+nfsd4_cb_getattr_release(struct nfsd4_callback *cb)
+{
+       struct nfs4_cb_fattr *ncf =
+                       container_of(cb, struct nfs4_cb_fattr, ncf_getattr);
+       struct nfs4_delegation *dp =
+                       container_of(ncf, struct nfs4_delegation, dl_cb_fattr);
+
+       nfs4_put_stid(&dp->dl_stid);
+       clear_bit(CB_GETATTR_BUSY, &ncf->ncf_cb_flags);
+       wake_up_bit(&ncf->ncf_cb_flags, CB_GETATTR_BUSY);
+}
+
 static const struct nfsd4_callback_ops nfsd4_cb_recall_any_ops = {
        .done           = nfsd4_cb_recall_any_done,
        .release        = nfsd4_cb_recall_any_release,
 };
 
+static const struct nfsd4_callback_ops nfsd4_cb_getattr_ops = {
+       .done           = nfsd4_cb_getattr_done,
+       .release        = nfsd4_cb_getattr_release,
+};
+
+static void nfs4_cb_getattr(struct nfs4_cb_fattr *ncf)
+{
+       struct nfs4_delegation *dp =
+                       container_of(ncf, struct nfs4_delegation, dl_cb_fattr);
+
+       if (test_and_set_bit(CB_GETATTR_BUSY, &ncf->ncf_cb_flags))
+               return;
+       /* set to proper status when nfsd4_cb_getattr_done runs */
+       ncf->ncf_cb_status = NFS4ERR_IO;
+
+       refcount_inc(&dp->dl_stid.sc_count);
+       nfsd4_run_cb(&ncf->ncf_getattr);
+}
+
 static struct nfs4_client *create_client(struct xdr_netobj name,
                struct svc_rqst *rqstp, nfs4_verifier *verf)
 {
@@ -5854,6 +5907,8 @@ nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp,
        struct svc_fh *parent = NULL;
        int cb_up;
        int status = 0;
+       struct kstat stat;
+       struct path path;
 
        cb_up = nfsd4_cb_channel_good(oo->oo_owner.so_client);
        open->op_recall = false;
@@ -5891,6 +5946,18 @@ nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp,
        if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) {
                open->op_delegate_type = NFS4_OPEN_DELEGATE_WRITE;
                trace_nfsd_deleg_write(&dp->dl_stid.sc_stateid);
+               path.mnt = currentfh->fh_export->ex_path.mnt;
+               path.dentry = currentfh->fh_dentry;
+               if (vfs_getattr(&path, &stat,
+                               (STATX_SIZE | STATX_CTIME | STATX_CHANGE_COOKIE),
+                               AT_STATX_SYNC_AS_STAT)) {
+                       nfs4_put_stid(&dp->dl_stid);
+                       destroy_delegation(dp);
+                       goto out_no_deleg;
+               }
+               dp->dl_cb_fattr.ncf_cur_fsize = stat.size;
+               dp->dl_cb_fattr.ncf_initial_cinfo =
+                       nfsd4_change_attribute(&stat, d_inode(currentfh->fh_dentry));
        } else {
                open->op_delegate_type = NFS4_OPEN_DELEGATE_READ;
                trace_nfsd_deleg_read(&dp->dl_stid.sc_stateid);
@@ -8710,6 +8777,8 @@ nfsd4_get_writestateid(struct nfsd4_compound_state *cstate,
  * nfsd4_deleg_getattr_conflict - Recall if GETATTR causes conflict
  * @rqstp: RPC transaction context
  * @inode: file to be checked for a conflict
+ * @modified: return true if file was modified
+ * @size: new size of file if modified is true
  *
  * This function is called when there is a conflict between a write
  * delegation and a change/size GETATTR from another client. The server
@@ -8718,22 +8787,22 @@ nfsd4_get_writestateid(struct nfsd4_compound_state *cstate,
  * delegation before replying to the GETATTR. See RFC 8881 section
  * 18.7.4.
  *
- * The current implementation does not support CB_GETATTR yet. However
- * this can avoid recalling the delegation could be added in follow up
- * work.
- *
  * Returns 0 if there is no conflict; otherwise an nfs_stat
  * code is returned.
  */
 __be32
-nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct inode *inode)
+nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct inode *inode,
+                               bool *modified, u64 *size)
 {
        __be32 status;
        struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
        struct file_lock_context *ctx;
        struct file_lock *fl;
        struct nfs4_delegation *dp;
+       struct iattr attrs;
+       struct nfs4_cb_fattr *ncf;
 
+       *modified = false;
        ctx = locks_inode_context(inode);
        if (!ctx)
                return 0;
@@ -8758,12 +8827,38 @@ nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct inode *inode)
                                return 0;
                        }
 break_lease:
-                       spin_unlock(&ctx->flc_lock);
                        nfsd_stats_wdeleg_getattr_inc(nn);
-                       status = nfserrno(nfsd_open_break_lease(inode, NFSD_MAY_READ));
-                       if (status != nfserr_jukebox ||
-                                       !nfsd_wait_for_delegreturn(rqstp, inode))
-                               return status;
+                       dp = fl->fl_owner;
+                       ncf = &dp->dl_cb_fattr;
+                       nfs4_cb_getattr(&dp->dl_cb_fattr);
+                       spin_unlock(&ctx->flc_lock);
+                       wait_on_bit_timeout(&ncf->ncf_cb_flags, CB_GETATTR_BUSY,
+                                       TASK_INTERRUPTIBLE, NFSD_CB_GETATTR_TIMEOUT);
+                       if (ncf->ncf_cb_status) {
+                               /* Recall delegation only if client didn't respond */
+                               status = nfserrno(nfsd_open_break_lease(inode, NFSD_MAY_READ));
+                               if (status != nfserr_jukebox ||
+                                               !nfsd_wait_for_delegreturn(rqstp, inode))
+                                       return status;
+                       }
+                       if (!ncf->ncf_file_modified &&
+                                       (ncf->ncf_initial_cinfo != ncf->ncf_cb_change ||
+                                       ncf->ncf_cur_fsize != ncf->ncf_cb_fsize))
+                               ncf->ncf_file_modified = true;
+                       if (ncf->ncf_file_modified) {
+                               /*
+                                * Per section 10.4.3 of RFC 8881, the server would
+                                * not update the file's metadata with the client's
+                                * modified size
+                                */
+                               attrs.ia_mtime = attrs.ia_ctime = current_time(inode);
+                               attrs.ia_valid = ATTR_MTIME | ATTR_CTIME;
+                               setattr_copy(&nop_mnt_idmap, inode, &attrs);
+                               mark_inode_dirty(inode);
+                               ncf->ncf_cur_fsize = ncf->ncf_cb_fsize;
+                               *size = ncf->ncf_cur_fsize;
+                               *modified = true;
+                       }
                        return 0;
                }
                break;
index e3f761cd5ee78d2847efd8bc50d2feab5ee9f9f2..9e8f230fc96e51ee28a4a1e5de06b4bfe325faa5 100644 (file)
@@ -3507,6 +3507,8 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr,
                unsigned long   mask[2];
        } u;
        unsigned long bit;
+       bool file_modified = false;
+       u64 size = 0;
 
        WARN_ON_ONCE(bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1);
        WARN_ON_ONCE(!nfsd_attrs_supported(minorversion, bmval));
@@ -3533,7 +3535,8 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr,
        }
        args.size = 0;
        if (u.attrmask[0] & (FATTR4_WORD0_CHANGE | FATTR4_WORD0_SIZE)) {
-               status = nfsd4_deleg_getattr_conflict(rqstp, d_inode(dentry));
+               status = nfsd4_deleg_getattr_conflict(rqstp, d_inode(dentry),
+                                       &file_modified, &size);
                if (status)
                        goto out;
        }
@@ -3543,7 +3546,10 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr,
                          AT_STATX_SYNC_AS_STAT);
        if (err)
                goto out_nfserr;
-       args.size = args.stat.size;
+       if (file_modified)
+               args.size = size;
+       else
+               args.size = args.stat.size;
 
        if (!(args.stat.result_mask & STATX_BTIME))
                /* underlying FS does not offer btime so we can't share it */
index 8daf22d766c60a9e20c74ab11c5d5ac63dec3312..16c5a05f340e51f761c69600fb9178efcd5d7689 100644 (file)
@@ -367,6 +367,7 @@ void                nfsd_lockd_shutdown(void);
 #define        NFSD_CLIENT_MAX_TRIM_PER_RUN    128
 #define        NFS4_CLIENTS_PER_GB             1024
 #define NFSD_DELEGRETURN_TIMEOUT       (HZ / 34)       /* 30ms */
+#define        NFSD_CB_GETATTR_TIMEOUT         NFSD_DELEGRETURN_TIMEOUT
 
 /*
  * The following attributes are currently not supported by the NFSv4 server:
index 3bf418ee6c970261820d64dc3465f5c5364593fe..01c6f344564693dd0987413c0ed116458782a677 100644 (file)
@@ -142,8 +142,16 @@ struct nfs4_cb_fattr {
        /* from CB_GETATTR reply */
        u64 ncf_cb_change;
        u64 ncf_cb_fsize;
+
+       unsigned long ncf_cb_flags;
+       bool ncf_file_modified;
+       u64 ncf_initial_cinfo;
+       u64 ncf_cur_fsize;
 };
 
+/* bits for ncf_cb_flags */
+#define        CB_GETATTR_BUSY         0
+
 /*
  * Represents a delegation stateid. The nfs4_client holds references to these
  * and they are put when it is being destroyed or when the delegation is
@@ -773,5 +781,5 @@ static inline bool try_to_expire_client(struct nfs4_client *clp)
 }
 
 extern __be32 nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp,
-                               struct inode *inode);
+               struct inode *inode, bool *file_modified, u64 *size);
 #endif   /* NFSD4_STATE_H */