Merge tag 'nfs-for-6.4-1' of git://git.linux-nfs.org/projects/anna/linux-nfs

Pull NFS client updates from Anna Schumaker:
 "New Features:

   - Convert the readdir path to use folios

   - Convert the NFS fscache code to use netfs

  Bugfixes and Cleanups:

   - Always send a RECLAIM_COMPLETE after establishing a lease

   - Simplify sysctl registrations and other cleanups

   - Handle out-of-order write replies on NFS v3

   - Have sunrpc call_bind_status use standard hard/soft task semantics

   - Other minor cleanups"

* tag 'nfs-for-6.4-1' of git://git.linux-nfs.org/projects/anna/linux-nfs:
  NFSv4.2: Rework scratch handling for READ_PLUS
  NFS: Cleanup unused rpc_clnt variable
  NFS: set varaiable nfs_netfs_debug_id storage-class-specifier to static
  SUNRPC: remove the maximum number of retries in call_bind_status
  NFS: Convert readdir page array functions to use a folio
  NFS: Convert the readdir array-of-pages into an array-of-folios
  NFSv3: handle out-of-order write replies.
  NFS: Remove fscache specific trace points and NFS_INO_FSCACHE bit
  NFS: Remove all NFSIOS_FSCACHE counters due to conversion to netfs API
  NFS: Convert buffered read paths to use netfs when fscache is enabled
  NFS: Configure support for netfs when NFS fscache is configured
  NFS: Rename readpage_async_filler to nfs_read_add_folio
  sunrpc: simplify one-level sysctl registration for debug_table
  sunrpc: move sunrpc_table and proc routines above
  sunrpc: simplify one-level sysctl registration for xs_tunables_table
  sunrpc: simplify one-level sysctl registration for xr_tunables_table
  nfs: simplify two-level sysctl registration for nfs_cb_sysctls
  nfs: simplify two-level sysctl registration for nfs4_cb_sysctls
  lockd: simplify two-level sysctl registration for nlm_sysctls
  NFSv4.1: Always send a RECLAIM_COMPLETE after establishing lease

1 files changed, 99 insertions, 15 deletions

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 97a76706fd54..a910b9a638c5 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -208,11 +208,12 @@ void nfs_set_cache_invalid(struct inode *inode, unsigned long flags)
 
 	nfsi->cache_validity |= flags;
 
-	if (inode->i_mapping->nrpages == 0)
-		nfsi->cache_validity &= ~(NFS_INO_INVALID_DATA |
-					  NFS_INO_DATA_INVAL_DEFER);
-	else if (nfsi->cache_validity & NFS_INO_INVALID_DATA)
-		nfsi->cache_validity &= ~NFS_INO_DATA_INVAL_DEFER;
+	if (inode->i_mapping->nrpages == 0) {
+		nfsi->cache_validity &= ~NFS_INO_INVALID_DATA;
+		nfs_ooo_clear(nfsi);
+	} else if (nfsi->cache_validity & NFS_INO_INVALID_DATA) {
+		nfs_ooo_clear(nfsi);
+	}
 	trace_nfs_set_cache_invalid(inode, 0);
 }
 EXPORT_SYMBOL_GPL(nfs_set_cache_invalid);
@@ -677,9 +678,10 @@ static int nfs_vmtruncate(struct inode * inode, loff_t offset)
 	trace_nfs_size_truncate(inode, offset);
 	i_size_write(inode, offset);
 	/* Optimisation */
-	if (offset == 0)
-		NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_DATA |
-				NFS_INO_DATA_INVAL_DEFER);
+	if (offset == 0) {
+		NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_DATA;
+		nfs_ooo_clear(NFS_I(inode));
+	}
 	NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE;
 
 	spin_unlock(&inode->i_lock);
@@ -1107,7 +1109,7 @@ void nfs_inode_attach_open_context(struct nfs_open_context *ctx)
 
 	spin_lock(&inode->i_lock);
 	if (list_empty(&nfsi->open_files) &&
-	    (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER))
+	    nfs_ooo_test(nfsi))
 		nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA |
 						     NFS_INO_REVAL_FORCED);
 	list_add_tail_rcu(&ctx->list, &nfsi->open_files);
@@ -1351,8 +1353,8 @@ int nfs_clear_invalid_mapping(struct address_space *mapping)
 
 	set_bit(NFS_INO_INVALIDATING, bitlock);
 	smp_wmb();
-	nfsi->cache_validity &=
-		~(NFS_INO_INVALID_DATA | NFS_INO_DATA_INVAL_DEFER);
+	nfsi->cache_validity &= ~NFS_INO_INVALID_DATA;
+	nfs_ooo_clear(nfsi);
 	spin_unlock(&inode->i_lock);
 	trace_nfs_invalidate_mapping_enter(inode);
 	ret = nfs_invalidate_mapping(inode, mapping);
@@ -1814,6 +1816,66 @@ static int nfs_inode_finish_partial_attr_update(const struct nfs_fattr *fattr,
 	return 0;
 }
 
+static void nfs_ooo_merge(struct nfs_inode *nfsi,
+			  u64 start, u64 end)
+{
+	int i, cnt;
+
+	if (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)
+		/* No point merging anything */
+		return;
+
+	if (!nfsi->ooo) {
+		nfsi->ooo = kmalloc(sizeof(*nfsi->ooo), GFP_ATOMIC);
+		if (!nfsi->ooo) {
+			nfsi->cache_validity |= NFS_INO_DATA_INVAL_DEFER;
+			return;
+		}
+		nfsi->ooo->cnt = 0;
+	}
+
+	/* add this range, merging if possible */
+	cnt = nfsi->ooo->cnt;
+	for (i = 0; i < cnt; i++) {
+		if (end == nfsi->ooo->gap[i].start)
+			end = nfsi->ooo->gap[i].end;
+		else if (start == nfsi->ooo->gap[i].end)
+			start = nfsi->ooo->gap[i].start;
+		else
+			continue;
+		/* Remove 'i' from table and loop to insert the new range */
+		cnt -= 1;
+		nfsi->ooo->gap[i] = nfsi->ooo->gap[cnt];
+		i = -1;
+	}
+	if (start != end) {
+		if (cnt >= ARRAY_SIZE(nfsi->ooo->gap)) {
+			nfsi->cache_validity |= NFS_INO_DATA_INVAL_DEFER;
+			kfree(nfsi->ooo);
+			nfsi->ooo = NULL;
+			return;
+		}
+		nfsi->ooo->gap[cnt].start = start;
+		nfsi->ooo->gap[cnt].end = end;
+		cnt += 1;
+	}
+	nfsi->ooo->cnt = cnt;
+}
+
+static void nfs_ooo_record(struct nfs_inode *nfsi,
+			   struct nfs_fattr *fattr)
+{
+	/* This reply was out-of-order, so record in the
+	 * pre/post change id, possibly cancelling
+	 * gaps created when iversion was jumpped forward.
+	 */
+	if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) &&
+	    (fattr->valid & NFS_ATTR_FATTR_PRECHANGE))
+		nfs_ooo_merge(nfsi,
+			      fattr->change_attr,
+			      fattr->pre_change_attr);
+}
+
 static int nfs_refresh_inode_locked(struct inode *inode,
 				    struct nfs_fattr *fattr)
 {
@@ -1824,8 +1886,12 @@ static int nfs_refresh_inode_locked(struct inode *inode,
 
 	if (attr_cmp > 0 || nfs_inode_finish_partial_attr_update(fattr, inode))
 		ret = nfs_update_inode(inode, fattr);
-	else if (attr_cmp == 0)
-		ret = nfs_check_inode_attributes(inode, fattr);
+	else {
+		nfs_ooo_record(NFS_I(inode), fattr);
+
+		if (attr_cmp == 0)
+			ret = nfs_check_inode_attributes(inode, fattr);
+	}
 
 	trace_nfs_refresh_inode_exit(inode, ret);
 	return ret;
@@ -1916,6 +1982,8 @@ int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fa
 	if (attr_cmp < 0)
 		return 0;
 	if ((fattr->valid & NFS_ATTR_FATTR) == 0 || !attr_cmp) {
+		/* Record the pre/post change info before clearing PRECHANGE */
+		nfs_ooo_record(NFS_I(inode), fattr);
 		fattr->valid &= ~(NFS_ATTR_FATTR_PRECHANGE
 				| NFS_ATTR_FATTR_PRESIZE
 				| NFS_ATTR_FATTR_PREMTIME
@@ -2070,6 +2138,15 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 
 	/* More cache consistency checks */
 	if (fattr->valid & NFS_ATTR_FATTR_CHANGE) {
+		if (!have_writers && nfsi->ooo && nfsi->ooo->cnt == 1 &&
+		    nfsi->ooo->gap[0].end == inode_peek_iversion_raw(inode)) {
+			/* There is one remaining gap that hasn't been
+			 * merged into iversion - do that now.
+			 */
+			inode_set_iversion_raw(inode, nfsi->ooo->gap[0].start);
+			kfree(nfsi->ooo);
+			nfsi->ooo = NULL;
+		}
 		if (!inode_eq_iversion_raw(inode, fattr->change_attr)) {
 			/* Could it be a race with writeback? */
 			if (!(have_writers || have_delegation)) {
@@ -2091,8 +2168,11 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 				dprintk("NFS: change_attr change on server for file %s/%ld\n",
 						inode->i_sb->s_id,
 						inode->i_ino);
-			} else if (!have_delegation)
-				nfsi->cache_validity |= NFS_INO_DATA_INVAL_DEFER;
+			} else if (!have_delegation) {
+				nfs_ooo_record(nfsi, fattr);
+				nfs_ooo_merge(nfsi, inode_peek_iversion_raw(inode),
+					      fattr->change_attr);
+			}
 			inode_set_iversion_raw(inode, fattr->change_attr);
 		}
 	} else {
@@ -2246,18 +2326,22 @@ struct inode *nfs_alloc_inode(struct super_block *sb)
 		return NULL;
 	nfsi->flags = 0UL;
 	nfsi->cache_validity = 0UL;
+	nfsi->ooo = NULL;
 #if IS_ENABLED(CONFIG_NFS_V4)
 	nfsi->nfs4_acl = NULL;
 #endif /* CONFIG_NFS_V4 */
 #ifdef CONFIG_NFS_V4_2
 	nfsi->xattr_cache = NULL;
 #endif
+	nfs_netfs_inode_init(nfsi);
+
 	return &nfsi->vfs_inode;
 }
 EXPORT_SYMBOL_GPL(nfs_alloc_inode);
 
 void nfs_free_inode(struct inode *inode)
 {
+	kfree(NFS_I(inode)->ooo);
 	kmem_cache_free(nfs_inode_cachep, NFS_I(inode));
 }
 EXPORT_SYMBOL_GPL(nfs_free_inode);