Merge tag '6.9-rc-smb3-client-fixes-part1' of git://git.samba.org/sfrench/cifs-2.6

Pull smb client updates from Steve French:

 - fix for folios/netfs data corruption in cifs_extend_writeback

 - additional tracepoint added

 - updates for special files and symlinks: improvements to allow
   selecting use of either WSL or NFS reparse point format on creating
   special files

 - allocation size improvement for cached files

 - minor cleanup patches

 - fix to allow changing the password on remount when password for the
   session is expired.

 - lease key related fixes: caching hardlinked files, deletes of
   deferred close files, and an important fix to better reuse lease keys
   for compound operations, which also can avoid lease break timeouts
   when low on credits

 - fix potential data corruption with write/readdir races

 - compression cleanups and a fix for compression headers

* tag '6.9-rc-smb3-client-fixes-part1' of git://git.samba.org/sfrench/cifs-2.6: (24 commits)
  cifs: update internal module version number for cifs.ko
  smb: common: simplify compression headers
  smb: common: fix fields sizes in compression_pattern_payload_v1
  smb: client: negotiate compression algorithms
  smb3: add dynamic trace point for ioctls
  cifs: Fix writeback data corruption
  smb: client: return reparse type in /proc/mounts
  smb: client: set correct d_type for reparse DFS/DFSR and mount point
  smb: client: parse uid, gid, mode and dev from WSL reparse points
  smb: client: introduce SMB2_OP_QUERY_WSL_EA
  smb: client: Fix a NULL vs IS_ERR() check in wsl_set_xattrs()
  smb: client: add support for WSL reparse points
  smb: client: reduce number of parameters in smb2_compound_op()
  smb: client: fix potential broken compound request
  smb: client: move most of reparse point handling code to common file
  smb: client: introduce reparse mount option
  smb: client: retry compound request without reusing lease
  smb: client: do not defer close open handles to deleted files
  smb: client: reuse file lease key in compound operations
  smb3: update allocation size more accurately on write completion
  ...

27 files changed, 1565 insertions, 707 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index 4bc35329837a..2696082d458e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5255,6 +5255,7 @@ R:	Paulo Alcantara <pc@manguebit.com> (DFS, global name space)
 R:	Ronnie Sahlberg <ronniesahlberg@gmail.com> (directory leases, sparse files)
 R:	Shyam Prasad N <sprasad@microsoft.com> (multichannel)
 R:	Tom Talpey <tom@talpey.com> (RDMA, smbdirect)
+R:	Bharath SM <bharathsm@microsoft.com> (deferred close, directory leases)
 L:	linux-cifs@vger.kernel.org
 L:	samba-technical@lists.samba.org (moderated for non-subscribers)
 S:	Supported
diff --git a/fs/smb/client/Makefile b/fs/smb/client/Makefile
index 0b07eb94c93b..e11985f2460b 100644
--- a/fs/smb/client/Makefile
+++ b/fs/smb/client/Makefile
@@ -12,7 +12,7 @@ cifs-y := trace.o cifsfs.o cifs_debug.o connect.o dir.o file.o \
 	  smb2ops.o smb2maperror.o smb2transport.o \
 	  smb2misc.o smb2pdu.o smb2inode.o smb2file.o cifsacl.o fs_context.o \
 	  dns_resolve.o cifs_spnego_negtokeninit.asn1.o asn1.o \
-	  namespace.o
+	  namespace.o reparse.o
 
 $(obj)/asn1.o: $(obj)/cifs_spnego_negtokeninit.asn1.h
 
diff --git a/fs/smb/client/cifs_debug.c b/fs/smb/client/cifs_debug.c
index 3e4209f41c18..226d4835c92d 100644
--- a/fs/smb/client/cifs_debug.c
+++ b/fs/smb/client/cifs_debug.c
@@ -278,6 +278,24 @@ static int cifs_debug_files_proc_show(struct seq_file *m, void *v)
 	return 0;
 }
 
+static __always_inline const char *compression_alg_str(__le16 alg)
+{
+	switch (alg) {
+	case SMB3_COMPRESS_NONE:
+		return "NONE";
+	case SMB3_COMPRESS_LZNT1:
+		return "LZNT1";
+	case SMB3_COMPRESS_LZ77:
+		return "LZ77";
+	case SMB3_COMPRESS_LZ77_HUFF:
+		return "LZ77-Huffman";
+	case SMB3_COMPRESS_PATTERN:
+		return "Pattern_V1";
+	default:
+		return "invalid";
+	}
+}
+
 static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 {
 	struct mid_q_entry *mid_entry;
@@ -423,12 +441,6 @@ skip_rdma:
 			server->echo_credits,
 			server->oplock_credits,
 			server->dialect);
-		if (server->compress_algorithm == SMB3_COMPRESS_LZNT1)
-			seq_printf(m, " COMPRESS_LZNT1");
-		else if (server->compress_algorithm == SMB3_COMPRESS_LZ77)
-			seq_printf(m, " COMPRESS_LZ77");
-		else if (server->compress_algorithm == SMB3_COMPRESS_LZ77_HUFF)
-			seq_printf(m, " COMPRESS_LZ77_HUFF");
 		if (server->sign)
 			seq_printf(m, " signed");
 		if (server->posix_ext_supported)
@@ -460,6 +472,14 @@ skip_rdma:
 				   server->leaf_fullpath);
 		}
 
+		seq_puts(m, "\nCompression: ");
+		if (!server->compression.requested)
+			seq_puts(m, "disabled on mount");
+		else if (server->compression.enabled)
+			seq_printf(m, "enabled (%s)", compression_alg_str(server->compression.alg));
+		else
+			seq_puts(m, "disabled (not supported by this server)");
+
 		seq_printf(m, "\n\n\tSessions: ");
 		i = 0;
 		list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) {
@@ -488,6 +508,8 @@ skip_rdma:
 				ses->ses_count, ses->serverOS, ses->serverNOS,
 				ses->capabilities, ses->ses_status);
 			}
+			if (ses->expired_pwd)
+				seq_puts(m, "password no longer valid ");
 			spin_unlock(&ses->ses_lock);
 
 			seq_printf(m, "\n\tSecurity type: %s ",
diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c
index e0d8c79cdde1..81d9aafd2210 100644
--- a/fs/smb/client/cifsfs.c
+++ b/fs/smb/client/cifsfs.c
@@ -673,6 +673,8 @@ cifs_show_options(struct seq_file *s, struct dentry *root)
 		seq_printf(s, ",backupgid=%u",
 			   from_kgid_munged(&init_user_ns,
 					    cifs_sb->ctx->backupgid));
+	seq_show_option(s, "reparse",
+			cifs_reparse_type_str(cifs_sb->ctx->reparse_type));
 
 	seq_printf(s, ",rsize=%u", cifs_sb->ctx->rsize);
 	seq_printf(s, ",wsize=%u", cifs_sb->ctx->wsize);
diff --git a/fs/smb/client/cifsfs.h b/fs/smb/client/cifsfs.h
index 685f7d1139c6..ca55d01117c8 100644
--- a/fs/smb/client/cifsfs.h
+++ b/fs/smb/client/cifsfs.h
@@ -152,6 +152,6 @@ extern const struct export_operations cifs_export_ops;
 #endif /* CONFIG_CIFS_NFSD_EXPORT */
 
 /* when changing internal version - update following two lines at same time */
-#define SMB3_PRODUCT_BUILD 47
-#define CIFS_VERSION   "2.47"
+#define SMB3_PRODUCT_BUILD 48
+#define CIFS_VERSION   "2.48"
 #endif				/* _CIFSFS_H */
diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h
index 53c75cfb33ab..8be62ed053a2 100644
--- a/fs/smb/client/cifsglob.h
+++ b/fs/smb/client/cifsglob.h
@@ -153,6 +153,24 @@ enum securityEnum {
 	Kerberos,		/* Kerberos via SPNEGO */
 };
 
+enum cifs_reparse_type {
+	CIFS_REPARSE_TYPE_NFS,
+	CIFS_REPARSE_TYPE_WSL,
+	CIFS_REPARSE_TYPE_DEFAULT = CIFS_REPARSE_TYPE_NFS,
+};
+
+static inline const char *cifs_reparse_type_str(enum cifs_reparse_type type)
+{
+	switch (type) {
+	case CIFS_REPARSE_TYPE_NFS:
+		return "nfs";
+	case CIFS_REPARSE_TYPE_WSL:
+		return "wsl";
+	default:
+		return "unknown";
+	}
+}
+
 struct session_key {
 	unsigned int len;
 	char *response;
@@ -208,6 +226,10 @@ struct cifs_open_info_data {
 			struct reparse_posix_data *posix;
 		};
 	} reparse;
+	struct {
+		__u8		eas[SMB2_WSL_MAX_QUERY_EA_RESP_SIZE];
+		unsigned int	eas_len;
+	} wsl;
 	char *symlink_target;
 	struct cifs_sid posix_owner;
 	struct cifs_sid posix_group;
@@ -217,19 +239,6 @@ struct cifs_open_info_data {
 	};
 };
 
-static inline bool cifs_open_data_reparse(struct cifs_open_info_data *data)
-{
-	struct smb2_file_all_info *fi = &data->fi;
-	u32 attrs = le32_to_cpu(fi->Attributes);
-	bool ret;
-
-	ret = data->reparse_point || (attrs & ATTR_REPARSE);
-	if (ret)
-		attrs |= ATTR_REPARSE;
-	fi->Attributes = cpu_to_le32(attrs);
-	return ret;
-}
-
 /*
  *****************************************************************
  * Except the CIFS PDUs themselves all the
@@ -371,7 +380,8 @@ struct smb_version_operations {
 			    struct cifs_open_info_data *data);
 	/* set size by path */
 	int (*set_path_size)(const unsigned int, struct cifs_tcon *,
-			     const char *, __u64, struct cifs_sb_info *, bool);
+			     const char *, __u64, struct cifs_sb_info *, bool,
+				 struct dentry *);
 	/* set size by file handle */
 	int (*set_file_size)(const unsigned int, struct cifs_tcon *,
 			     struct cifsFileInfo *, __u64, bool);
@@ -401,7 +411,7 @@ struct smb_version_operations {
 		     struct cifs_sb_info *);
 	/* unlink file */
 	int (*unlink)(const unsigned int, struct cifs_tcon *, const char *,
-		      struct cifs_sb_info *);
+		      struct cifs_sb_info *, struct dentry *);
 	/* open, rename and delete file */
 	int (*rename_pending_delete)(const char *, struct dentry *,
 				     const unsigned int);
@@ -759,7 +769,11 @@ struct TCP_Server_Info {
 	unsigned int	max_write;
 	unsigned int	min_offload;
 	unsigned int	retrans;
-	__le16	compress_algorithm;
+	struct {
+		bool requested; /* "compress" mount option set*/
+		bool enabled; /* actually negotiated with server */
+		__le16 alg; /* preferred alg negotiated with server */
+	} compression;
 	__u16	signing_algorithm;
 	__le16	cipher_type;
 	 /* save initital negprot hash */
@@ -1066,6 +1080,7 @@ struct cifs_ses {
 	enum securityEnum sectype; /* what security flavor was specified? */
 	bool sign;		/* is signing required? */
 	bool domainAuto:1;
+	bool expired_pwd;  /* track if access denied or expired pwd so can know if need to update */
 	unsigned int flags;
 	__u16 session_flags;
 	__u8 smb3signingkey[SMB3_SIGN_KEY_SIZE];
@@ -1379,6 +1394,7 @@ struct cifs_open_parms {
 	umode_t mode;
 	bool reconnect:1;
 	bool replay:1; /* indicates that this open is for a replay */
+	struct kvec *ea_cctx;
 };
 
 struct cifs_fid {
@@ -1420,6 +1436,7 @@ struct cifsFileInfo {
 	bool invalidHandle:1;	/* file closed via session abend */
 	bool swapfile:1;
 	bool oplock_break_cancelled:1;
+	bool status_file_deleted:1; /* file has been deleted */
 	unsigned int oplock_epoch; /* epoch from the lease break */
 	__u32 oplock_level; /* oplock/lease level from the lease break */
 	int count;
@@ -2277,6 +2294,17 @@ static inline void cifs_sg_set_buf(struct sg_table *sgtable,
 	}
 }
 
+#define CIFS_OPARMS(_cifs_sb, _tcon, _path, _da, _cd, _co, _mode) \
+	((struct cifs_open_parms) { \
+		.tcon = _tcon, \
+		.path = _path, \
+		.desired_access = (_da), \
+		.disposition = (_cd), \
+		.create_options = cifs_create_options(_cifs_sb, (_co)), \
+		.mode = (_mode), \
+		.cifs_sb = _cifs_sb, \
+	})
+
 struct smb2_compound_vars {
 	struct cifs_open_parms oparms;
 	struct kvec rsp_iov[MAX_COMPOUND];
@@ -2288,6 +2316,7 @@ struct smb2_compound_vars {
 	struct kvec close_iov;
 	struct smb2_file_rename_info rename_info;
 	struct smb2_file_link_info link_info;
+	struct kvec ea_iov;
 };
 
 #endif	/* _CIFS_GLOB_H */
diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h
index a841bf4967fa..0723e1b57256 100644
--- a/fs/smb/client/cifsproto.h
+++ b/fs/smb/client/cifsproto.h
@@ -144,7 +144,8 @@ extern int cifs_reconnect(struct TCP_Server_Info *server,
 extern int checkSMB(char *buf, unsigned int len, struct TCP_Server_Info *srvr);
 extern bool is_valid_oplock_break(char *, struct TCP_Server_Info *);
 extern bool backup_cred(struct cifs_sb_info *);
-extern bool is_size_safe_to_change(struct cifsInodeInfo *, __u64 eof);
+extern bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 eof,
+				   bool from_readdir);
 extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset,
 			    unsigned int bytes_written);
 extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, int);
@@ -201,17 +202,14 @@ extern void cifs_unix_basic_to_fattr(struct cifs_fattr *fattr,
 				     struct cifs_sb_info *cifs_sb);
 extern void cifs_dir_info_to_fattr(struct cifs_fattr *, FILE_DIRECTORY_INFO *,
 					struct cifs_sb_info *);
-extern int cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr);
+extern int cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr,
+			       bool from_readdir);
 extern struct inode *cifs_iget(struct super_block *sb,
 			       struct cifs_fattr *fattr);
 
 int cifs_get_inode_info(struct inode **inode, const char *full_path,
 			struct cifs_open_info_data *data, struct super_block *sb, int xid,
 			const struct cifs_fid *fid);
-bool cifs_reparse_point_to_fattr(struct cifs_sb_info *cifs_sb,
-				 struct cifs_fattr *fattr,
-				 struct cifs_open_info_data *data);
-
 extern int smb311_posix_get_inode_info(struct inode **inode,
 				       const char *full_path,
 				       struct cifs_open_info_data *data,
@@ -296,6 +294,10 @@ extern void cifs_close_all_deferred_files(struct cifs_tcon *cifs_tcon);
 
 extern void cifs_close_deferred_file_under_dentry(struct cifs_tcon *cifs_tcon,
 				const char *path);
+
+extern void cifs_mark_open_handles_for_deleted_file(struct inode *inode,
+				const char *path);
+
 extern struct TCP_Server_Info *
 cifs_get_tcp_session(struct smb3_fs_context *ctx,
 		     struct TCP_Server_Info *primary_server);
@@ -402,7 +404,8 @@ extern int CIFSSMBSetFileDisposition(const unsigned int xid,
 				     __u32 pid_of_opener);
 extern int CIFSSMBSetEOF(const unsigned int xid, struct cifs_tcon *tcon,
 			 const char *file_name, __u64 size,
-			 struct cifs_sb_info *cifs_sb, bool set_allocation);
+			 struct cifs_sb_info *cifs_sb, bool set_allocation,
+			 struct dentry *dentry);
 extern int CIFSSMBSetFileSize(const unsigned int xid, struct cifs_tcon *tcon,
 			      struct cifsFileInfo *cfile, __u64 size,
 			      bool set_allocation);
@@ -438,7 +441,8 @@ extern int CIFSPOSIXDelFile(const unsigned int xid, struct cifs_tcon *tcon,
 			const struct nls_table *nls_codepage,
 			int remap_special_chars);
 extern int CIFSSMBDelFile(const unsigned int xid, struct cifs_tcon *tcon,
-			  const char *name, struct cifs_sb_info *cifs_sb);
+			  const char *name, struct cifs_sb_info *cifs_sb,
+			  struct dentry *dentry);
 int CIFSSMBRename(const unsigned int xid, struct cifs_tcon *tcon,
 		  struct dentry *source_dentry,
 		  const char *from_name, const char *to_name,
diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c
index 5eb83bafc7fd..5aee55551573 100644
--- a/fs/smb/client/cifssmb.c
+++ b/fs/smb/client/cifssmb.c
@@ -738,7 +738,7 @@ PsxDelete:
 
 int
 CIFSSMBDelFile(const unsigned int xid, struct cifs_tcon *tcon, const char *name,
-	       struct cifs_sb_info *cifs_sb)
+	       struct cifs_sb_info *cifs_sb, struct dentry *dentry)
 {
 	DELETE_FILE_REQ *pSMB = NULL;
 	DELETE_FILE_RSP *pSMBr = NULL;
@@ -4993,7 +4993,7 @@ QFSPosixRetry:
 int
 CIFSSMBSetEOF(const unsigned int xid, struct cifs_tcon *tcon,
 	      const char *file_name, __u64 size, struct cifs_sb_info *cifs_sb,
-	      bool set_allocation)
+	      bool set_allocation, struct dentry *dentry)
 {
 	struct smb_com_transaction2_spi_req *pSMB = NULL;
 	struct smb_com_transaction2_spi_rsp *pSMBr = NULL;
diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c
index ac9595504f4b..86ae578904a2 100644
--- a/fs/smb/client/connect.c
+++ b/fs/smb/client/connect.c
@@ -1736,7 +1736,7 @@ cifs_get_tcp_session(struct smb3_fs_context *ctx,
 	tcp_ses->channel_sequence_num = 0; /* only tracked for primary channel */
 	tcp_ses->reconnect_instance = 1;
 	tcp_ses->lstrp = jiffies;
-	tcp_ses->compress_algorithm = cpu_to_le16(ctx->compression);
+	tcp_ses->compression.requested = ctx->compress;
 	spin_lock_init(&tcp_ses->req_lock);
 	spin_lock_init(&tcp_ses->srv_lock);
 	spin_lock_init(&tcp_ses->mid_lock);
@@ -2803,6 +2803,8 @@ compare_mount_options(struct super_block *sb, struct cifs_mnt_data *mnt_data)
 		return 0;
 	if (old->ctx->closetimeo != new->ctx->closetimeo)
 		return 0;
+	if (old->ctx->reparse_type != new->ctx->reparse_type)
+		return 0;
 
 	return 1;
 }
diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c
index c3b8e7091a4d..ec25d3c3e1ee 100644
--- a/fs/smb/client/file.c
+++ b/fs/smb/client/file.c
@@ -329,7 +329,7 @@ int cifs_posix_open(const char *full_path, struct inode **pinode,
 		}
 	} else {
 		cifs_revalidate_mapping(*pinode);
-		rc = cifs_fattr_to_inode(*pinode, &fattr);
+		rc = cifs_fattr_to_inode(*pinode, &fattr, false);
 	}
 
 posix_open_ret:
@@ -486,6 +486,7 @@ struct cifsFileInfo *cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 	cfile->uid = current_fsuid();
 	cfile->dentry = dget(dentry);
 	cfile->f_flags = file->f_flags;
+	cfile->status_file_deleted = false;
 	cfile->invalidHandle = false;
 	cfile->deferred_close_scheduled = false;
 	cfile->tlink = cifs_get_tlink(tlink);
@@ -1088,7 +1089,7 @@ int cifs_close(struct inode *inode, struct file *file)
 		if ((cifs_sb->ctx->closetimeo && cinode->oplock == CIFS_CACHE_RHW_FLG)
 		    && cinode->lease_granted &&
 		    !test_bit(CIFS_INO_CLOSE_ON_LOCK, &cinode->flags) &&
-		    dclose) {
+		    dclose && !(cfile->status_file_deleted)) {
 			if (test_and_clear_bit(CIFS_INO_MODIFIED_ATTR, &cinode->flags)) {
 				inode_set_mtime_to_ts(inode,
 						      inode_set_ctime_current(inode));
@@ -2628,20 +2629,20 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to)
  * dirty pages if possible, but don't sleep while doing so.
  */
 static void cifs_extend_writeback(struct address_space *mapping,
+				  struct xa_state *xas,
 				  long *_count,
 				  loff_t start,
 				  int max_pages,
-				  size_t max_len,
-				  unsigned int *_len)
+				  loff_t max_len,
+				  size_t *_len)
 {
 	struct folio_batch batch;
 	struct folio *folio;
-	unsigned int psize, nr_pages;
-	size_t len = *_len;
-	pgoff_t index = (start + len) / PAGE_SIZE;
+	unsigned int nr_pages;
+	pgoff_t index = (start + *_len) / PAGE_SIZE;
+	size_t len;
 	bool stop = true;
 	unsigned int i;
-	XA_STATE(xas, &mapping->i_pages, index);
 
 	folio_batch_init(&batch);
 
@@ -2652,54 +2653,64 @@ static void cifs_extend_writeback(struct address_space *mapping,
 		 */
 		rcu_read_lock();
 
-		xas_for_each(&xas, folio, ULONG_MAX) {
+		xas_for_each(xas, folio, ULONG_MAX) {
 			stop = true;
-			if (xas_retry(&xas, folio))
+			if (xas_retry(xas, folio))
 				continue;
 			if (xa_is_value(folio))
 				break;
-			if (folio->index != index)
+			if (folio->index != index) {
+				xas_reset(xas);
 				break;
+			}
+
 			if (!folio_try_get_rcu(folio)) {
-				xas_reset(&xas);
+				xas_reset(xas);
 				continue;
 			}
 			nr_pages = folio_nr_pages(folio);
-			if (nr_pages > max_pages)
+			if (nr_pages > max_pages) {
+				xas_reset(xas);
 				break;
+			}
 
 			/* Has the page moved or been split? */
-			if (unlikely(folio != xas_reload(&xas))) {
+			if (unlikely(folio != xas_reload(xas))) {
 				folio_put(folio);
+				xas_reset(xas);
 				break;
 			}
 
 			if (!folio_trylock(folio)) {
 				folio_put(folio);
+				xas_reset(xas);
 				break;
 			}
-			if (!folio_test_dirty(folio) || folio_test_writeback(folio)) {
+			if (!folio_test_dirty(folio) ||
+			    folio_test_writeback(folio)) {
 				folio_unlock(folio);
 				folio_put(folio);
+				xas_reset(xas);
 				break;
 			}
 
 			max_pages -= nr_pages;
-			psize = folio_size(folio);
-			len += psize;
+			len = folio_size(folio);
 			stop = false;
-			if (max_pages <= 0 || len >= max_len || *_count <= 0)
-				stop = true;
 
 			index += nr_pages;
+			*_count -= nr_pages;
+			*_len += len;
+			if (max_pages <= 0 || *_len >= max_len || *_count <= 0)
+				stop = true;
+
 			if (!folio_batch_add(&batch, folio))
 				break;
 			if (stop)
 				break;
 		}
 
-		if (!stop)
-			xas_pause(&xas);
+		xas_pause(xas);
 		rcu_read_unlock();
 
 		/* Now, if we obtained any pages, we can shift them to being
@@ -2716,16 +2727,12 @@ static void cifs_extend_writeback(struct address_space *mapping,
 			if (!folio_clear_dirty_for_io(folio))
 				WARN_ON(1);
 			folio_start_writeback(folio);
-
-			*_count -= folio_nr_pages(folio);
 			folio_unlock(folio);
 		}
 
 		folio_batch_release(&batch);
 		cond_resched();
 	} while (!stop);
-
-	*_len = len;
 }
 
 /*
@@ -2733,8 +2740,10 @@ static void cifs_extend_writeback(struct address_space *mapping,
  */
 static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping,
 						 struct writeback_control *wbc,
+						 struct xa_state *xas,
 						 struct folio *folio,
-						 loff_t start, loff_t end)
+						 unsigned long long start,
+						 unsigned long long end)
 {
 	struct inode *inode = mapping->host;
 	struct TCP_Server_Info *server;
@@ -2743,17 +2752,18 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping,
 	struct cifs_credits credits_on_stack;
 	struct cifs_credits *credits = &credits_on_stack;
 	struct cifsFileInfo *cfile = NULL;
-	unsigned int xid, wsize, len;
-	loff_t i_size = i_size_read(inode);
-	size_t max_len;
+	unsigned long long i_size = i_size_read(inode), max_len;
+	unsigned int xid, wsize;
+	size_t len = folio_size(folio);
 	long count = wbc->nr_to_write;
 	int rc;
 
 	/* The folio should be locked, dirty and not undergoing writeback. */
+	if (!folio_clear_dirty_for_io(folio))
+		WARN_ON_ONCE(1);
 	folio_start_writeback(folio);
 
 	count -= folio_nr_pages(folio);
-	len = folio_size(folio);
 
 	xid = get_xid();
 	server = cifs_pick_channel(cifs_sb_master_tcon(cifs_sb)->ses);
@@ -2783,9 +2793,10 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping,
 	wdata->server = server;
 	cfile = NULL;
 
-	/* Find all consecutive lockable dirty pages, stopping when we find a
-	 * page that is not immediately lockable, is not dirty or is missing,
-	 * or we reach the end of the range.
+	/* Find all consecutive lockable dirty pages that have contiguous
+	 * written regions, stopping when we find a page that is not
+	 * immediately lockable, is not dirty or is missing, or we reach the
+	 * end of the range.
 	 */
 	if (start < i_size) {
 		/* Trim the write to the EOF; the extra data is ignored.  Also
@@ -2805,19 +2816,18 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping,
 			max_pages -= folio_nr_pages(folio);
 
 			if (max_pages > 0)
-				cifs_extend_writeback(mapping, &count, start,
+				cifs_extend_writeback(mapping, xas, &count, start,
 						      max_pages, max_len, &len);
 		}
-		len = min_t(loff_t, len, max_len);
 	}
-
-	wdata->bytes = len;
+	len = min_t(unsigned long long, len, i_size - start);
 
 	/* We now have a contiguous set of dirty pages, each with writeback
 	 * set; the first page is still locked at this point, but all the rest
 	 * have been unlocked.
 	 */
 	folio_unlock(folio);
+	wdata->bytes = len;
 
 	if (start < i_size) {
 		iov_iter_xarray(&wdata->iter, ITER_SOURCE, &mapping->i_pages,
@@ -2868,102 +2878,118 @@ err_xid:
 /*
  * write a region of pages back to the server
  */
-static int cifs_writepages_region(struct address_space *mapping,
-				  struct writeback_control *wbc,
-				  loff_t start, loff_t end, loff_t *_next)
+static ssize_t cifs_writepages_begin(struct address_space *mapping,
+				     struct writeback_control *wbc,
+				     struct xa_state *xas,
+				     unsigned long long *_start,
+				     unsigned long long end)
 {
-	struct folio_batch fbatch;
+	struct folio *folio;
+	unsigned long long start = *_start;
+	ssize_t ret;
 	int skips = 0;
 
-	folio_batch_init(&fbatch);
-	do {
-		int nr;
-		pgoff_t index = start / PAGE_SIZE;
+search_again:
+	/* Find the first dirty page. */
+	rcu_read_lock();
 
-		nr = filemap_get_folios_tag(mapping, &index, end / PAGE_SIZE,
-					    PAGECACHE_TAG_DIRTY, &fbatch);
-		if (!nr)
+	for (;;) {
+		folio = xas_find_marked(xas, end / PAGE_SIZE, PAGECACHE_TAG_DIRTY);
+		if (xas_retry(xas, folio) || xa_is_value(folio))
+			continue;
+		if (!folio)
 			break;
 
-		for (int i = 0; i < nr; i++) {
-			ssize_t ret;
-			struct folio *folio = fbatch.folios[i];
+		if (!folio_try_get_rcu(folio)) {
+			xas_reset(xas);
+			continue;
+		}
 
-redo_folio:
-			start = folio_pos(folio); /* May regress with THPs */
+		if (unlikely(folio != xas_reload(xas))) {
+			folio_put(folio);
+			xas_reset(xas);
+			continue;
+		}
 
-			/* At this point we hold neither the i_pages lock nor the
-			 * page lock: the page may be truncated or invalidated
-			 * (changing page->mapping to NULL), or even swizzled
-			 * back from swapper_space to tmpfs file mapping
-			 */
-			if (wbc->sync_mode != WB_SYNC_NONE) {
-				ret = folio_lock_killable(folio);
-				if (ret < 0)
-					goto write_error;
-			} else {
-				if (!folio_trylock(folio))
-					goto skip_write;
-			}
+		xas_pause(xas);
+		break;
+	}
+	rcu_read_unlock();
+	if (!folio)
+		return 0;
 
-			if (folio->mapping != mapping ||
-			    !folio_test_dirty(folio)) {
-				start += folio_size(folio);
-				folio_unlock(folio);
-				continue;
-			}
+	start = folio_pos(folio); /* May regress with THPs */
 
-			if (folio_test_writeback(folio) ||
-			    folio_test_fscache(folio)) {
-				folio_unlock(folio);
-				if (wbc->sync_mode == WB_SYNC_NONE)
-					goto skip_write;
+	/* At this point we hold neither the i_pages lock nor the page lock:
+	 * the page may be truncated or invalidated (changing page->mapping to
+	 * NULL), or even swizzled back from swapper_space to tmpfs file
+	 * mapping
+	 */
+lock_again:
+	if (wbc->sync_mode != WB_SYNC_NONE) {
+		ret = folio_lock_killable(folio);
+		if (ret < 0)
+			return ret;
+	} else {
+		if (!folio_trylock(folio))
+			goto search_again;
+	}
+
+	if (folio->mapping != mapping ||
+	    !folio_test_dirty(folio)) {
+		start += folio_size(folio);
+		folio_unlock(folio);
+		goto search_again;
+	}
 
-				folio_wait_writeback(folio);
+	if (folio_test_writeback(folio) ||
+	    folio_test_fscache(folio)) {
+		folio_unlock(folio);
+		if (wbc->sync_mode != WB_SYNC_NONE) {
+			folio_wait_writeback(folio);
 #ifdef CONFIG_CIFS_FSCACHE
-				folio_wait_fscache(folio);
+			folio_wait_fscache(folio);
 #endif
-				goto redo_folio;
-			}
-
-			if (!folio_clear_dirty_for_io(folio))
-				/* We hold the page lock - it should've been dirty. */
-				WARN_ON(1);
-
-			ret = cifs_write_back_from_locked_folio(mapping, wbc, folio, start, end);
-			if (ret < 0)
-				goto write_error;
-
-			start += ret;
-			continue;
-
-write_error:
-			folio_batch_release(&fbatch);
-			*_next = start;
-			return ret;
+			goto lock_again;
+		}
 
-skip_write:
-			/*
-			 * Too many skipped writes, or need to reschedule?
-			 * Treat it as a write error without an error code.
-			 */
+		start += folio_size(folio);
+		if (wbc->sync_mode == WB_SYNC_NONE) {
 			if (skips >= 5 || need_resched()) {
 				ret = 0;
-				goto write_error;
+				goto out;
 			}
-
-			/* Otherwise, just skip that folio and go on to the next */
 			skips++;
-			start += folio_size(folio);
-			continue;
 		}
+		goto search_again;
+	}
 
-		folio_batch_release(&fbatch);
-		cond_resched();
-	} while (wbc->nr_to_write > 0);
+	ret = cifs_write_back_from_locked_folio(mapping, wbc, xas, folio, start, end);
+out:
+	if (ret > 0)
+		*_start = start + ret;
+	return ret;
+}
 
-	*_next = start;
-	return 0;
+/*
+ * Write a region of pages back to the server
+ */
+static int cifs_writepages_region(struct address_space *mapping,
+				  struct writeback_control *wbc,
+				  unsigned long long *_start,
+				  unsigned long long end)
+{
+	ssize_t ret;
+
+	XA_STATE(xas, &mapping->i_pages, *_start / PAGE_SIZE);
+
+	do {
+		ret = cifs_writepages_begin(mapping, wbc, &xas, _start, end);
+		if (ret > 0 && wbc->nr_to_write > 0)
+			cond_resched();
+	} while (ret > 0 && wbc->nr_to_write > 0);
+
+	return ret > 0 ? 0 : ret;
 }
 
 /*
@@ -2972,7 +2998,7 @@ skip_write:
 static int cifs_writepages(struct address_space *mapping,
 			   struct writeback_control *wbc)
 {
-	loff_t start, next;
+	loff_t start, end;
 	int ret;
 
 	/* We have to be careful as we can end up racing with setattr()
@@ -2980,28 +3006,34 @@ static int cifs_writepages(struct address_space *mapping,
 	 * to prevent it.
 	 */
 
-	if (wbc->range_cyclic) {
+	if (wbc->range_cyclic && mapping->writeback_index) {
 		start = mapping->writeback_index * PAGE_SIZE;
-		ret = cifs_writepages_region(mapping, wbc, start, LLONG_MAX, &next);
-		if (ret == 0) {
-			mapping->writeback_ind