1 files changed, 150 insertions, 30 deletions

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index acef52ecb1bb..119e447758b9 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -74,6 +74,8 @@ nfs_fattr_to_ino_t(struct nfs_fattr *fattr)
 
 int nfs_wait_bit_killable(struct wait_bit_key *key, int mode)
 {
+	if (unlikely(nfs_current_task_exiting()))
+		return -EINTR;
 	schedule();
 	if (signal_pending_state(mode, current))
 		return -ERESTARTSYS;
@@ -190,9 +192,8 @@ static bool nfs_has_xattr_cache(const struct nfs_inode *nfsi)
 void nfs_set_cache_invalid(struct inode *inode, unsigned long flags)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
-	bool have_delegation = NFS_PROTO(inode)->have_delegation(inode, FMODE_READ);
 
-	if (have_delegation) {
+	if (nfs_have_delegated_attributes(inode)) {
 		if (!(flags & NFS_INO_REVAL_FORCED))
 			flags &= ~(NFS_INO_INVALID_MODE |
 				   NFS_INO_INVALID_OTHER |
@@ -206,12 +207,15 @@ void nfs_set_cache_invalid(struct inode *inode, unsigned long flags)
 		nfs_fscache_invalidate(inode, 0);
 	flags &= ~NFS_INO_REVAL_FORCED;
 
-	nfsi->cache_validity |= flags;
+	flags |= nfsi->cache_validity;
+	if (inode->i_mapping->nrpages == 0)
+		flags &= ~NFS_INO_INVALID_DATA;
 
-	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) {
+	/* pairs with nfs_clear_invalid_mapping()'s smp_load_acquire() */
+	smp_store_release(&nfsi->cache_validity, flags);
+
+	if (inode->i_mapping->nrpages == 0 ||
+	    nfsi->cache_validity & NFS_INO_INVALID_DATA) {
 		nfs_ooo_clear(nfsi);
 	}
 	trace_nfs_set_cache_invalid(inode, 0);
@@ -276,6 +280,8 @@ EXPORT_SYMBOL_GPL(nfs_zap_acl_cache);
 
 void nfs_invalidate_atime(struct inode *inode)
 {
+	if (nfs_have_delegated_atime(inode))
+		return;
 	spin_lock(&inode->i_lock);
 	nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME);
 	spin_unlock(&inode->i_lock);
@@ -491,6 +497,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
 			inode->i_fop = NFS_SB(sb)->nfs_client->rpc_ops->file_ops;
 			inode->i_data.a_ops = &nfs_file_aops;
 			nfs_inode_init_regular(nfsi);
+			mapping_set_large_folios(inode->i_mapping);
 		} else if (S_ISDIR(inode->i_mode)) {
 			inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->dir_inode_ops;
 			inode->i_fop = &nfs_dir_operations;
@@ -604,6 +611,67 @@ out_no_inode:
 }
 EXPORT_SYMBOL_GPL(nfs_fhget);
 
+static void
+nfs_fattr_fixup_delegated(struct inode *inode, struct nfs_fattr *fattr)
+{
+	unsigned long cache_validity = NFS_I(inode)->cache_validity;
+
+	if (nfs_have_delegated_mtime(inode)) {
+		if (!(cache_validity & NFS_INO_INVALID_CTIME))
+			fattr->valid &= ~(NFS_ATTR_FATTR_PRECTIME |
+					  NFS_ATTR_FATTR_CTIME);
+
+		if (!(cache_validity & NFS_INO_INVALID_MTIME))
+			fattr->valid &= ~(NFS_ATTR_FATTR_PREMTIME |
+					  NFS_ATTR_FATTR_MTIME);
+
+		if (!(cache_validity & NFS_INO_INVALID_ATIME))
+			fattr->valid &= ~NFS_ATTR_FATTR_ATIME;
+	} else if (nfs_have_delegated_atime(inode)) {
+		if (!(cache_validity & NFS_INO_INVALID_ATIME))
+			fattr->valid &= ~NFS_ATTR_FATTR_ATIME;
+	}
+}
+
+static void nfs_update_timestamps(struct inode *inode, unsigned int ia_valid)
+{
+	enum file_time_flags time_flags = 0;
+	unsigned int cache_flags = 0;
+
+	if (ia_valid & ATTR_MTIME) {
+		time_flags |= S_MTIME | S_CTIME;
+		cache_flags |= NFS_INO_INVALID_CTIME | NFS_INO_INVALID_MTIME;
+	}
+	if (ia_valid & ATTR_ATIME) {
+		time_flags |= S_ATIME;
+		cache_flags |= NFS_INO_INVALID_ATIME;
+	}
+	inode_update_timestamps(inode, time_flags);
+	NFS_I(inode)->cache_validity &= ~cache_flags;
+}
+
+void nfs_update_delegated_atime(struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	if (nfs_have_delegated_atime(inode))
+		nfs_update_timestamps(inode, ATTR_ATIME);
+	spin_unlock(&inode->i_lock);
+}
+
+void nfs_update_delegated_mtime_locked(struct inode *inode)
+{
+	if (nfs_have_delegated_mtime(inode))
+		nfs_update_timestamps(inode, ATTR_MTIME);
+}
+
+void nfs_update_delegated_mtime(struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	nfs_update_delegated_mtime_locked(inode);
+	spin_unlock(&inode->i_lock);
+}
+EXPORT_SYMBOL_GPL(nfs_update_delegated_mtime);
+
 #define NFS_VALID_ATTRS (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_SIZE|ATTR_ATIME|ATTR_ATIME_SET|ATTR_MTIME|ATTR_MTIME_SET|ATTR_FILE|ATTR_OPEN)
 
 int
@@ -631,6 +699,18 @@ nfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
 			attr->ia_valid &= ~ATTR_SIZE;
 	}
 
+	if (nfs_have_delegated_mtime(inode) && attr->ia_valid & ATTR_MTIME) {
+		spin_lock(&inode->i_lock);
+		nfs_update_timestamps(inode, attr->ia_valid);
+		spin_unlock(&inode->i_lock);
+		attr->ia_valid &= ~(ATTR_MTIME | ATTR_ATIME);
+	} else if (nfs_have_delegated_atime(inode) &&
+		   attr->ia_valid & ATTR_ATIME &&
+		   !(attr->ia_valid & ATTR_MTIME)) {
+		nfs_update_delegated_atime(inode);
+		attr->ia_valid &= ~ATTR_ATIME;
+	}
+
 	/* Optimization: if the end result is no change, don't RPC */
 	if (((attr->ia_valid & NFS_VALID_ATTRS) & ~(ATTR_FILE|ATTR_OPEN)) == 0)
 		return 0;
@@ -686,6 +766,7 @@ static int nfs_vmtruncate(struct inode * inode, loff_t offset)
 
 	spin_unlock(&inode->i_lock);
 	truncate_pagecache(inode, offset);
+	nfs_update_delegated_mtime_locked(inode);
 	spin_lock(&inode->i_lock);
 out:
 	return err;
@@ -709,8 +790,9 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
 	spin_lock(&inode->i_lock);
 	NFS_I(inode)->attr_gencount = fattr->gencount;
 	if ((attr->ia_valid & ATTR_SIZE) != 0) {
-		nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME |
-						     NFS_INO_INVALID_BLOCKS);
+		if (!nfs_have_delegated_mtime(inode))
+			nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME);
+		nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS);
 		nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC);
 		nfs_vmtruncate(inode, attr->ia_size);
 	}
@@ -856,8 +938,12 @@ int nfs_getattr(struct mnt_idmap *idmap, const struct path *path,
 
 	/* Flush out writes to the server in order to update c/mtime/version.  */
 	if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_CHANGE_COOKIE)) &&
-	    S_ISREG(inode->i_mode))
-		filemap_write_and_wait(inode->i_mapping);
+	    S_ISREG(inode->i_mode)) {
+		if (nfs_have_delegated_mtime(inode))
+			filemap_fdatawrite(inode->i_mapping);
+		else
+			filemap_write_and_wait(inode->i_mapping);
+	}
 
 	/*
 	 * We may force a getattr if the user cares about atime.
@@ -1012,7 +1098,7 @@ void nfs_close_context(struct nfs_open_context *ctx, int is_sync)
 	if (!is_sync)
 		return;
 	inode = d_inode(ctx->dentry);
-	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
+	if (nfs_have_read_or_write_delegation(inode))
 		return;
 	nfsi = NFS_I(inode);
 	if (inode->i_mapping->nrpages == 0)
@@ -1053,6 +1139,8 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry,
 	ctx->lock_context.open_context = ctx;
 	INIT_LIST_HEAD(&ctx->list);
 	ctx->mdsthreshold = NULL;
+	nfs_localio_file_init(&ctx->nfl);
+
 	return ctx;
 }
 EXPORT_SYMBOL_GPL(alloc_nfs_open_context);
@@ -1084,6 +1172,7 @@ static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync)
 	nfs_sb_deactive(sb);
 	put_rpccred(rcu_dereference_protected(ctx->ll_cred, 1));
 	kfree(ctx->mdsthreshold);
+	nfs_close_local_fh(&ctx->nfl);
 	kfree_rcu(ctx, rcu_head);
 }
 
@@ -1340,6 +1429,13 @@ int nfs_clear_invalid_mapping(struct address_space *mapping)
 					 TASK_KILLABLE|TASK_FREEZABLE_UNSAFE);
 		if (ret)
 			goto out;
+		smp_rmb(); /* pairs with smp_wmb() below */
+		if (test_bit(NFS_INO_INVALIDATING, bitlock))
+			continue;
+		/* pairs with nfs_set_cache_invalid()'s smp_store_release() */
+		if (!(smp_load_acquire(&nfsi->cache_validity) & NFS_INO_INVALID_DATA))
+			goto out;
+		/* Slow-path that double-checks with spinlock held */
 		spin_lock(&inode->i_lock);
 		if (test_bit(NFS_INO_INVALIDATING, bitlock)) {
 			spin_unlock(&inode->i_lock);
@@ -1482,7 +1578,7 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat
 	unsigned long invalid = 0;
 	struct timespec64 ts;
 
-	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
+	if (nfs_have_delegated_attributes(inode))
 		return 0;
 
 	if (!(fattr->valid & NFS_ATTR_FATTR_FILEID)) {
@@ -1565,6 +1661,7 @@ void nfs_fattr_init(struct nfs_fattr *fattr)
 	fattr->gencount = nfs_inc_attr_generation_counter();
 	fattr->owner_name = NULL;
 	fattr->group_name = NULL;
+	fattr->mdsthreshold = NULL;
 }
 EXPORT_SYMBOL_GPL(nfs_fattr_init);
 
@@ -2118,6 +2215,9 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 	 */
 	nfsi->read_cache_jiffies = fattr->time_start;
 
+	/* Fix up any delegated attributes in the struct nfs_fattr */
+	nfs_fattr_fixup_delegated(inode, fattr);
+
 	save_cache_validity = nfsi->cache_validity;
 	nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR
 			| NFS_INO_INVALID_ATIME
@@ -2390,35 +2490,54 @@ static void nfs_destroy_inodecache(void)
 	kmem_cache_destroy(nfs_inode_cachep);
 }
 
+struct workqueue_struct *nfslocaliod_workqueue;
 struct workqueue_struct *nfsiod_workqueue;
 EXPORT_SYMBOL_GPL(nfsiod_workqueue);
 
 /*
- * start up the nfsiod workqueue
+ * Destroy the nfsiod workqueues
  */
-static int nfsiod_start(void)
+static void nfsiod_stop(void)
 {
 	struct workqueue_struct *wq;
-	dprintk("RPC:       creating workqueue nfsiod\n");
-	wq = alloc_workqueue("nfsiod", WQ_MEM_RECLAIM | WQ_UNBOUND, 0);
-	if (wq == NULL)
-		return -ENOMEM;
-	nfsiod_workqueue = wq;
-	return 0;
+
+	wq = nfsiod_workqueue;
+	if (wq != NULL) {
+		nfsiod_workqueue = NULL;
+		destroy_workqueue(wq);
+	}
+#if IS_ENABLED(CONFIG_NFS_LOCALIO)
+	wq = nfslocaliod_workqueue;
+	if (wq != NULL) {
+		nfslocaliod_workqueue = NULL;
+		destroy_workqueue(wq);
+	}
+#endif /* CONFIG_NFS_LOCALIO */
 }
 
 /*
- * Destroy the nfsiod workqueue
+ * Start the nfsiod workqueues
  */
-static void nfsiod_stop(void)
+static int nfsiod_start(void)
 {
-	struct workqueue_struct *wq;
-
-	wq = nfsiod_workqueue;
-	if (wq == NULL)
-		return;
-	nfsiod_workqueue = NULL;
-	destroy_workqueue(wq);
+	dprintk("RPC:       creating workqueue nfsiod\n");
+	nfsiod_workqueue = alloc_workqueue("nfsiod", WQ_MEM_RECLAIM | WQ_UNBOUND, 0);
+	if (nfsiod_workqueue == NULL)
+		return -ENOMEM;
+#if IS_ENABLED(CONFIG_NFS_LOCALIO)
+	/*
+	 * localio writes need to use a normal (non-memreclaim) workqueue.
+	 * When we start getting low on space, XFS goes and calls flush_work() on
+	 * a non-memreclaim work queue, which causes a priority inversion problem.
+	 */
+	dprintk("RPC:       creating workqueue nfslocaliod\n");
+	nfslocaliod_workqueue = alloc_workqueue("nfslocaliod", WQ_UNBOUND, 0);
+	if (unlikely(nfslocaliod_workqueue == NULL)) {
+		nfsiod_stop();
+		return -ENOMEM;
+	}
+#endif /* CONFIG_NFS_LOCALIO */
+	return 0;
 }
 
 unsigned int nfs_net_id;
@@ -2538,6 +2657,7 @@ static void __exit exit_nfs_fs(void)
 
 /* Not quite true; I just maintain it */
 MODULE_AUTHOR("Olaf Kirch <okir@monad.swb.de>");
+MODULE_DESCRIPTION("NFS client support");
 MODULE_LICENSE("GPL");
 module_param(enable_ino64, bool, 0644);