Merge tag 'nfs-for-6.17-3' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client fixes from Trond Myklebust:
"Stable patches:

- Revert "SUNRPC: Don't allow waiting for exiting tasks" as it is
breaking ltp tests

Bugfixes:

- Another set of fixes to the tracking of NFSv4 server capabilities
when crossing filesystem boundaries

- Localio fix to restore credentials and prevent triggering a
BUG_ON()

- Fix to prevent flapping of the localio on/off trigger

- Protections against 'eof page pollution' as demonstrated in
xfstests generic/363

- Series of patches to ensure correct ordering of O_DIRECT i/o and
truncate, fallocate and copy functions

- Fix a NULL pointer check in flexfiles reads that regresses 6.17

- Correct a typo that breaks flexfiles layout segment processing"

* tag 'nfs-for-6.17-3' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
NFSv4/flexfiles: Fix layout merge mirror check.
SUNRPC: call xs_sock_process_cmsg for all cmsg
Revert "SUNRPC: Don't allow waiting for exiting tasks"
NFS: Fix the marking of the folio as up to date
NFS: nfs_invalidate_folio() must observe the offset and size arguments
NFSv4.2: Serialise O_DIRECT i/o and copy range
NFSv4.2: Serialise O_DIRECT i/o and clone range
NFSv4.2: Serialise O_DIRECT i/o and fallocate()
NFS: Serialise O_DIRECT i/o and truncate()
NFSv4.2: Protect copy offload and clone against 'eof page pollution'
NFS: Protect against 'eof page pollution'
flexfiles/pNFS: fix NULL checks on result of ff_layout_choose_ds_for_read
nfs/localio: avoid bouncing LOCALIO if nfs_client_is_local()
nfs/localio: restore creds before releasing pageio data
NFSv4: Clear the NFS_CAP_XATTR flag if not supported by the server
NFSv4: Clear NFS_CAP_OPEN_XOR and NFS_CAP_DELEGTIME if not supported
NFSv4: Clear the NFS_CAP_FS_LOCATIONS flag if it is not set
NFSv4: Don't clear capabilities that won't be reset

fs/nfs/client.c

···
 
 	if (fsinfo->xattr_support)
 		server->caps |= NFS_CAP_XATTR;
+	else
+		server->caps &= ~NFS_CAP_XATTR;
 #endif
 }
 

fs/nfs/file.c

···
 #include <linux/mm.h>
 #include <linux/pagemap.h>
 #include <linux/gfp.h>
+#include <linux/rmap.h>
 #include <linux/swap.h>
 #include <linux/compaction.h>
 
···
 }
 EXPORT_SYMBOL_GPL(nfs_file_fsync);
 
+void nfs_truncate_last_folio(struct address_space *mapping, loff_t from,
+			     loff_t to)
+{
+	struct folio *folio;
+
+	if (from >= to)
+		return;
+
+	folio = filemap_lock_folio(mapping, from >> PAGE_SHIFT);
+	if (IS_ERR(folio))
+		return;
+
+	if (folio_mkclean(folio))
+		folio_mark_dirty(folio);
+
+	if (folio_test_uptodate(folio)) {
+		loff_t fpos = folio_pos(folio);
+		size_t offset = from - fpos;
+		size_t end = folio_size(folio);
+
+		if (to - fpos < end)
+			end = to - fpos;
+		folio_zero_segment(folio, offset, end);
+		trace_nfs_size_truncate_folio(mapping->host, to);
+	}
+
+	folio_unlock(folio);
+	folio_put(folio);
+}
+EXPORT_SYMBOL_GPL(nfs_truncate_last_folio);
+
 /*
  * Decide whether a read/modify/write cycle may be more efficient
  * then a modify/write/read cycle when writing to a page in the
···
 
 	dfprintk(PAGECACHE, "NFS: write_begin(%pD2(%lu), %u@%lld)\n",
 		file, mapping->host->i_ino, len, (long long) pos);
+	nfs_truncate_last_folio(mapping, i_size_read(mapping->host), pos);
 
 	fgp |= fgf_set_order(len);
 start:
···
 	dfprintk(PAGECACHE, "NFS: invalidate_folio(%lu, %zu, %zu)\n",
 		 folio->index, offset, length);
 
-	if (offset != 0 || length < folio_size(folio))
-		return;
 	/* Cancel any unstarted writes on this page */
-	nfs_wb_folio_cancel(inode, folio);
+	if (offset != 0 || length < folio_size(folio))
+		nfs_wb_folio(inode, folio);
+	else
+		nfs_wb_folio_cancel(inode, folio);
 	folio_wait_private_2(folio); /* [DEPRECATED] */
 	trace_nfs_invalidate_folio(inode, folio_pos(folio) + offset, length);
 }

fs/nfs/flexfilelayout/flexfilelayout.c

···
 		struct pnfs_layout_segment *l2)
 {
 	const struct nfs4_ff_layout_segment *fl1 = FF_LAYOUT_LSEG(l1);
-	const struct nfs4_ff_layout_segment *fl2 = FF_LAYOUT_LSEG(l1);
+	const struct nfs4_ff_layout_segment *fl2 = FF_LAYOUT_LSEG(l2);
 	u32 i;
 
 	if (fl1->mirror_array_cnt != fl2->mirror_array_cnt)
···
 			continue;
 
 		if (check_device &&
-		    nfs4_test_deviceid_unavailable(&mirror->mirror_ds->id_node))
+		    nfs4_test_deviceid_unavailable(&mirror->mirror_ds->id_node)) {
+			// reinitialize the error state in case if this is the last iteration
+			ds = ERR_PTR(-EINVAL);
 			continue;
+		}
 
 		*best_idx = idx;
 		break;
···
 	struct nfs4_pnfs_ds *ds;
 
 	ds = ff_layout_choose_valid_ds_for_read(lseg, start_idx, best_idx);
-	if (ds)
+	if (!IS_ERR(ds))
 		return ds;
 	return ff_layout_choose_any_ds_for_read(lseg, start_idx, best_idx);
 }
···
 
 	ds = ff_layout_choose_best_ds_for_read(lseg, pgio->pg_mirror_idx,
 					       best_idx);
-	if (ds || !pgio->pg_mirror_idx)
+	if (!IS_ERR(ds) || !pgio->pg_mirror_idx)
 		return ds;
 	return ff_layout_choose_best_ds_for_read(lseg, 0, best_idx);
 }
···
 	req->wb_nio = 0;
 
 	ds = ff_layout_get_ds_for_read(pgio, &ds_idx);
-	if (!ds) {
+	if (IS_ERR(ds)) {
 		if (!ff_layout_no_fallback_to_mds(pgio->pg_lseg))
 			goto out_mds;
 		pnfs_generic_pg_cleanup(pgio);
···
 {
 	u32 idx = hdr->pgio_mirror_idx + 1;
 	u32 new_idx = 0;
+	struct nfs4_pnfs_ds *ds;
 
-	if (ff_layout_choose_any_ds_for_read(hdr->lseg, idx, &new_idx))
-		ff_layout_send_layouterror(hdr->lseg);
-	else
+	ds = ff_layout_choose_any_ds_for_read(hdr->lseg, idx, &new_idx);
+	if (IS_ERR(ds))
 		pnfs_error_mark_layout_for_return(hdr->inode, hdr->lseg);
+	else
+		ff_layout_send_layouterror(hdr->lseg);
 	pnfs_read_resend_pnfs(hdr, new_idx);
 }
 

fs/nfs/inode.c

···
 {
 	struct inode *inode = d_inode(dentry);
 	struct nfs_fattr *fattr;
+	loff_t oldsize = i_size_read(inode);
 	int error = 0;
 
 	nfs_inc_stats(inode, NFSIOS_VFSSETATTR);
···
 		if (error)
 			return error;
 
-		if (attr->ia_size == i_size_read(inode))
+		if (attr->ia_size == oldsize)
 			attr->ia_valid &= ~ATTR_SIZE;
 	}
 
···
 	trace_nfs_setattr_enter(inode);
 
 	/* Write all dirty data */
-	if (S_ISREG(inode->i_mode))
+	if (S_ISREG(inode->i_mode)) {
+		nfs_file_block_o_direct(NFS_I(inode));
 		nfs_sync_inode(inode);
+	}
 
 	fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode));
 	if (fattr == NULL) {
···
 	}
 
 	error = NFS_PROTO(inode)->setattr(dentry, fattr, attr);
-	if (error == 0)
+	if (error == 0) {
+		if (attr->ia_valid & ATTR_SIZE)
+			nfs_truncate_last_folio(inode->i_mapping, oldsize,
+						attr->ia_size);
 		error = nfs_refresh_inode(inode, fattr);
+	}
 	nfs_free_fattr(fattr);
 out:
 	trace_nfs_setattr_exit(inode, error);

fs/nfs/internal.h

···
 int nfs_lock(struct file *, int, struct file_lock *);
 int nfs_flock(struct file *, int, struct file_lock *);
 int nfs_check_flags(int);
+void nfs_truncate_last_folio(struct address_space *mapping, loff_t from,
+			     loff_t to);
 
 /* inode.c */
 extern struct workqueue_struct *nfsiod_workqueue;
···
 {
 	return test_bit(NFS_INO_ODIRECT, &nfsi->flags) == 0;
 }
+
+/* Must be called with exclusively locked inode->i_rwsem */
+static inline void nfs_file_block_o_direct(struct nfs_inode *nfsi)
+{
+	if (test_bit(NFS_INO_ODIRECT, &nfsi->flags)) {
+		clear_bit(NFS_INO_ODIRECT, &nfsi->flags);
+		inode_dio_wait(&nfsi->vfs_inode);
+	}
+}
+
 
 /* namespace.c */
 #define NFS_PATH_CANONICAL 1

fs/nfs/io.c

···
 
 #include "internal.h"
 
-/* Call with exclusively locked inode->i_rwsem */
-static void nfs_block_o_direct(struct nfs_inode *nfsi, struct inode *inode)
-{
-	if (test_bit(NFS_INO_ODIRECT, &nfsi->flags)) {
-		clear_bit(NFS_INO_ODIRECT, &nfsi->flags);
-		inode_dio_wait(inode);
-	}
-}
-
 /**
  * nfs_start_io_read - declare the file is being used for buffered reads
  * @inode: file inode
···
 	err = down_write_killable(&inode->i_rwsem);
 	if (err)
 		return err;
-	nfs_block_o_direct(nfsi, inode);
+	nfs_file_block_o_direct(nfsi);
 	downgrade_write(&inode->i_rwsem);
 
 	return 0;
···
 
 	err = down_write_killable(&inode->i_rwsem);
 	if (!err)
-		nfs_block_o_direct(NFS_I(inode), inode);
+		nfs_file_block_o_direct(NFS_I(inode));
 	return err;
 }
 

fs/nfs/localio.c

···
 		return;
 	}
 
-	if (nfs_client_is_local(clp)) {
-		/* If already enabled, disable and re-enable */
-		nfs_localio_disable_client(clp);
-	}
+	if (nfs_client_is_local(clp))
+		return;
 
 	if (!nfs_uuid_begin(&clp->cl_uuid))
 		return;
···
 		case -ENOMEM:
 		case -ENXIO:
 		case -ENOENT:
-			/* Revalidate localio, will disable if unsupported */
+			/* Revalidate localio */
+			nfs_localio_disable_client(clp);
 			nfs_local_probe(clp);
 		}
 	}
···
 	nfs_local_iter_init(&iter, iocb, READ);
 
 	status = filp->f_op->read_iter(&iocb->kiocb, &iter);
+
+	revert_creds(save_cred);
+
 	if (status != -EIOCBQUEUED) {
 		nfs_local_read_done(iocb, status);
 		nfs_local_pgio_release(iocb);
 	}
-
-	revert_creds(save_cred);
 }
 
 static int
···
 	file_start_write(filp);
 	status = filp->f_op->write_iter(&iocb->kiocb, &iter);
 	file_end_write(filp);
+
+	revert_creds(save_cred);
+	current->flags = old_flags;
+
 	if (status != -EIOCBQUEUED) {
 		nfs_local_write_done(iocb, status);
 		nfs_local_vfs_getattr(iocb);
 		nfs_local_pgio_release(iocb);
 	}
-
-	revert_creds(save_cred);
-	current->flags = old_flags;
 }
 
 static int

fs/nfs/nfs42proc.c

···
 	exception.inode = inode;
 	exception.state = lock->open_context->state;
 
+	nfs_file_block_o_direct(NFS_I(inode));
 	err = nfs_sync_inode(inode);
 	if (err)
 		goto out;
···
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ALLOCATE],
 	};
 	struct inode *inode = file_inode(filep);
+	loff_t oldsize = i_size_read(inode);
 	int err;
 
 	if (!nfs_server_capable(inode, NFS_CAP_ALLOCATE))
···
 	inode_lock(inode);
 
 	err = nfs42_proc_fallocate(&msg, filep, offset, len);
-	if (err == -EOPNOTSUPP)
+
+	if (err == 0)
+		nfs_truncate_last_folio(inode->i_mapping, oldsize,
+					offset + len);
+	else if (err == -EOPNOTSUPP)
 		NFS_SERVER(inode)->caps &= ~(NFS_CAP_ALLOCATE |
 					     NFS_CAP_ZERO_RANGE);
 
···
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ZERO_RANGE],
 	};
 	struct inode *inode = file_inode(filep);
+	loff_t oldsize = i_size_read(inode);
 	int err;
 
 	if (!nfs_server_capable(inode, NFS_CAP_ZERO_RANGE))
···
 	inode_lock(inode);
 
 	err = nfs42_proc_fallocate(&msg, filep, offset, len);
-	if (err == 0)
+	if (err == 0) {
+		nfs_truncate_last_folio(inode->i_mapping, oldsize,
+					offset + len);
 		truncate_pagecache_range(inode, offset, (offset + len) -1);
-	if (err == -EOPNOTSUPP)
+	} else if (err == -EOPNOTSUPP)
 		NFS_SERVER(inode)->caps &= ~NFS_CAP_ZERO_RANGE;
 
 	inode_unlock(inode);
···
 
 /**
  * nfs42_copy_dest_done - perform inode cache updates after clone/copy offload
- * @inode: pointer to destination inode
+ * @file: pointer to destination file
  * @pos: destination offset
  * @len: copy length
+ * @oldsize: length of the file prior to clone/copy
  *
  * Punch a hole in the inode page cache, so that the NFS client will
  * know to retrieve new data.
  * Update the file size if necessary, and then mark the inode as having
  * invalid cached values for change attribute, ctime, mtime and space used.
  */
-static void nfs42_copy_dest_done(struct inode *inode, loff_t pos, loff_t len)
+static void nfs42_copy_dest_done(struct file *file, loff_t pos, loff_t len,
+				 loff_t oldsize)
 {
+	struct inode *inode = file_inode(file);
+	struct address_space *mapping = file->f_mapping;
 	loff_t newsize = pos + len;
 	loff_t end = newsize - 1;
 
-	WARN_ON_ONCE(invalidate_inode_pages2_range(inode->i_mapping,
-				pos >> PAGE_SHIFT, end >> PAGE_SHIFT));
+	nfs_truncate_last_folio(mapping, oldsize, pos);
+	WARN_ON_ONCE(invalidate_inode_pages2_range(mapping, pos >> PAGE_SHIFT,
+						   end >> PAGE_SHIFT));
 
 	spin_lock(&inode->i_lock);
 	if (newsize > i_size_read(inode))
···
 	struct nfs_server *src_server = NFS_SERVER(src_inode);
 	loff_t pos_src = args->src_pos;
 	loff_t pos_dst = args->dst_pos;
+	loff_t oldsize_dst = i_size_read(dst_inode);
 	size_t count = args->count;
 	ssize_t status;
 
···
 		return status;
 	}
 
+	nfs_file_block_o_direct(NFS_I(dst_inode));
 	status = nfs_sync_inode(dst_inode);
 	if (status)
 		return status;
···
 			goto out;
 	}
 
-	nfs42_copy_dest_done(dst_inode, pos_dst, res->write_res.count);
+	nfs42_copy_dest_done(dst, pos_dst, res->write_res.count, oldsize_dst);
 	nfs_invalidate_atime(src_inode);
 	status = res->write_res.count;
 out:
···
 	struct nfs42_clone_res res = {
 		.server	= server,
 	};
+	loff_t oldsize_dst = i_size_read(dst_inode);
 	int status;
 
 	msg->rpc_argp = &args;
···
 		/* a zero-length count means clone to EOF in src */
 		if (count == 0 && res.dst_fattr->valid & NFS_ATTR_FATTR_SIZE)
 			count = nfs_size_to_loff_t(res.dst_fattr->size) - dst_offset;
-		nfs42_copy_dest_done(dst_inode, dst_offset, count);
+		nfs42_copy_dest_done(dst_f, dst_offset, count, oldsize_dst);
 		status = nfs_post_op_update_inode(dst_inode, res.dst_fattr);
 	}
 

fs/nfs/nfs4file.c

···
 	lock_two_nondirectories(src_inode, dst_inode);
 	/* flush all pending writes on both src and dst so that server
 	 * has the latest data */
+	nfs_file_block_o_direct(NFS_I(src_inode));
 	ret = nfs_sync_inode(src_inode);
 	if (ret)
 		goto out_unlock;
+	nfs_file_block_o_direct(NFS_I(dst_inode));
 	ret = nfs_sync_inode(dst_inode);
 	if (ret)
 		goto out_unlock;

fs/nfs/nfs4proc.c

···
 				     res.attr_bitmask[2];
 		}
 		memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask));
-		server->caps &= ~(NFS_CAP_ACLS | NFS_CAP_HARDLINKS |
-				  NFS_CAP_SYMLINKS| NFS_CAP_SECURITY_LABEL);
+		server->caps &=
+			~(NFS_CAP_ACLS | NFS_CAP_HARDLINKS | NFS_CAP_SYMLINKS |
+			  NFS_CAP_SECURITY_LABEL | NFS_CAP_FS_LOCATIONS |
+			  NFS_CAP_OPEN_XOR | NFS_CAP_DELEGTIME);
 		server->fattr_valid = NFS_ATTR_FATTR_V4;
 		if (res.attr_bitmask[0] & FATTR4_WORD0_ACL &&
 				res.acl_bitmask & ACL4_SUPPORT_ALLOW_ACL)
···
 	};
 	int err;
 
-	nfs_server_set_init_caps(server);
 	do {
 		err = nfs4_handle_exception(server,
 				_nfs4_server_capabilities(server, fhandle),

fs/nfs/nfstrace.h

···
 			TP_ARGS(inode, new_size))
 
 DEFINE_NFS_UPDATE_SIZE_EVENT(truncate);
+DEFINE_NFS_UPDATE_SIZE_EVENT(truncate_folio);
 DEFINE_NFS_UPDATE_SIZE_EVENT(wcc);
 DEFINE_NFS_UPDATE_SIZE_EVENT(update);
 DEFINE_NFS_UPDATE_SIZE_EVENT(grow);

fs/nfs/write.c

···
 }
 
 /*
- * nfs_page_group_search_locked
- * @head - head request of page group
- * @page_offset - offset into page
+ * nfs_page_covers_folio
+ * @req: struct nfs_page
  *
- * Search page group with head @head to find a request that contains the
- * page offset @page_offset.
- *
- * Returns a pointer to the first matching nfs request, or NULL if no
- * match is found.
- *
- * Must be called with the page group lock held
- */
-static struct nfs_page *
-nfs_page_group_search_locked(struct nfs_page *head, unsigned int page_offset)
-{
-	struct nfs_page *req;
-
-	req = head;
-	do {
-		if (page_offset >= req->wb_pgbase &&
-		    page_offset < (req->wb_pgbase + req->wb_bytes))
-			return req;
-
-		req = req->wb_this_page;
-	} while (req != head);
-
-	return NULL;
-}
-
-/*
- * nfs_page_group_covers_page
- * @head - head request of page group
- *
- * Return true if the page group with head @head covers the whole page,
- * returns false otherwise
+ * Return true if the request covers the whole folio.
+ * Note that the caller should ensure all subrequests have been joined
  */
 static bool nfs_page_group_covers_page(struct nfs_page *req)
 {
 	unsigned int len = nfs_folio_length(nfs_page_to_folio(req));
-	struct nfs_page *tmp;
-	unsigned int pos = 0;
 
-	nfs_page_group_lock(req);
-
-	for (;;) {
-		tmp = nfs_page_group_search_locked(req->wb_head, pos);
-		if (!tmp)
-			break;
-		pos = tmp->wb_pgbase + tmp->wb_bytes;
-	}
-
-	nfs_page_group_unlock(req);
-	return pos >= len;
+	return req->wb_pgbase == 0 && req->wb_bytes == len;
 }
 
 /* We can set the PG_uptodate flag if we see that a write request
···
 		 * release it */
 		nfs_inode_remove_request(req);
 		nfs_unlock_and_release_request(req);
+		folio_cancel_dirty(folio);
 	}
 
 	return ret;

net/sunrpc/sched.c

···
 
 static int rpc_wait_bit_killable(struct wait_bit_key *key, int mode)
 {
-	if (unlikely(current->flags & PF_EXITING))
-		return -EINTR;
 	schedule();
 	if (signal_pending_state(mode, current))
 		return -ERESTARTSYS;

net/sunrpc/xprtsock.c

···
 	iov_iter_kvec(&msg.msg_iter, ITER_DEST, &alert_kvec, 1,
 		      alert_kvec.iov_len);
 	ret = sock_recvmsg(sock, &msg, flags);
-	if (ret > 0 &&
-	    tls_get_record_type(sock->sk, &u.cmsg) == TLS_RECORD_TYPE_ALERT) {
-		iov_iter_revert(&msg.msg_iter, ret);
+	if (ret > 0) {
+		if (tls_get_record_type(sock->sk, &u.cmsg) == TLS_RECORD_TYPE_ALERT)
+			iov_iter_revert(&msg.msg_iter, ret);
 		ret = xs_sock_process_cmsg(sock, &msg, msg_flags, &u.cmsg,
 					   -EAGAIN);
 	}