Skip to contentSkip to conversation

netfs, cifs: Delegate high-level I/O to netfslib

PATCH Superseded David Howells 3 participants 23 messages Feb 5, 2024 – Feb 20, 2024

Subsystems (8)

v5 of 7 Superseded Latest is v7 →

lore
Share & tools
lore

David Howells

author maintainer
#0 lore raw

v5 cover netfs, cifs: Delegate high-level I/O to netfslib

Hi Steve,

Here are patches to convert cifs to use the netfslib library. With this I can run a certain amount of xfstests on CIFS, though not all the tests work correctly because of fallocate issues.

The patches remove around 2000 lines from CIFS

To do:

(*) Implement write-retry. Currently, netfslib errors out on a failed DIO
write and relies on the VM to drive writepages again on a failed buffered write. This needs some retry logic adding into fs/netfs/output.c.

I'm not sure what the best way to handle this is. One way is to resend each failing subreq as it fails, offloading this to a kernel thread that re-splits the subreq, calling out to a rreq->op to do the splitting, thereby allowing cifs to renegotiate credits. If a subreq is split, the two parts need to be adjacent in the rreq->subrequests list.

An alternative way might be to try and combine failing tests and then split them.

Yet a third way might be to try each failing subreq a smaller bit at a time and keep track of what has been sent in wdata->subreq.transferred.

Whichever way is chosen, NETFS_SREQ_RETRYING should be set in wdata->subreq.flags instead of setting wdata->replay.

Notes:

(1) CIFS is made to use unbuffered I/O for unbuffered caching modes and write-through caching for cache=strict.

(2) Various cifs fallocate() function implementations have issues that aren't easily fixed without enhanced protocol support.

(3) It should be possible to turn on multipage folio support in CIFS now.

(4) The then-unused CIFS code is removed in three patches, not one, to avoid the git patch generator from producing confusing patches in which it thinks code is being moved around rather than just being removed.

The patches can be found here also:

https://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git/log/?h=cifs-netfs

Changes
=======
ver #5)
- Rebased to -rc3 plus SteveF's for-next branch as netfslib is now upstream, as are a couple of patches from this series.
- Replace the ->replay bool Shyam added with a flag on the netfs subrequest. This is tested by the code, but not currently set (see above).

ver #4)
- Slimmed down the branch:
- Split the cifs-related patches off to a separate branch (cifs-netfs)
- Deferred the content-encryption to the in-progress ceph changes.
- Deferred the use-PG_writeback rather than PG_fscache patch
- Rebased on a later linux-next with afs-rotation patches.

ver #3)
- Moved the fscache module into netfslib to avoid export cycles.
- Fixed a bunch of bugs.
- Got CIFS to pass as much of xfstests as possible.
- Added a patch to make 9P use all the helpers.
- Added a patch to stop using PG_fscache, but rather dirty pages on reading and have writepages write to the cache.

ver #2)
- Folded the addition of NETFS_RREQ_NONBLOCK/BLOCKED into first patch that uses them.
- Folded addition of rsize member into first user.
- Don't set rsize in ceph (yet) and set it in kafs to 256KiB. cifs sets it dynamically.
- Moved direct_bv next to direct_bv_count in struct netfs_io_request and labelled it with a __counted_by().
- Passed flags into netfs_xa_store_and_mark() rather than two bools.
- Removed netfs_set_up_buffer() as it wasn't used.

David

David Howells (12):
cifs: Replace cifs_readdata with a wrapper around netfs_io_subrequest
cifs: Set zero_point in the copy_file_range() and remap_file_range()
cifs: Replace cifs_writedata with a wrapper around netfs_io_subrequest
cifs: Use more fields from netfs_io_subrequest
cifs: Make wait_mtu_credits take size_t args
cifs: Implement netfslib hooks
cifs: Replace the writedata replay bool with a netfs sreq flag
cifs: Move cifs_loose_read_iter() and cifs_file_write_iter() to file.c
cifs: Cut over to using netfslib
cifs: Remove some code that's no longer used, part 1
cifs: Remove some code that's no longer used, part 2
cifs: Remove some code that's no longer used, part 3

 fs/netfs/buffered_write.c    |    3 +
 fs/netfs/io.c                |    7 +-
 fs/smb/client/Kconfig        |    1 +
 fs/smb/client/cifsfs.c       |   69 +-
 fs/smb/client/cifsfs.h       |   10 +-
 fs/smb/client/cifsglob.h     |   59 +-
 fs/smb/client/cifsproto.h    |   14 +-
 fs/smb/client/cifssmb.c      |  111 +-
 fs/smb/client/file.c         | 2911 ++++++----------------------------
 fs/smb/client/fscache.c      |  109 --
 fs/smb/client/fscache.h      |   54 -
 fs/smb/client/inode.c        |   19 +-
 fs/smb/client/smb2ops.c      |   10 +-
 fs/smb/client/smb2pdu.c      |  169 +-
 fs/smb/client/smb2proto.h    |    5 +-
 fs/smb/client/trace.h        |  144 +-
 fs/smb/client/transport.c    |   17 +-
 include/linux/netfs.h        |    2 +
 include/trace/events/netfs.h |    1 +
 19 files changed, 852 insertions(+), 2863 deletions(-)
author maintainer
#1 lore raw

PATCH v5 1/12 cifs: Replace cifs_readdata with a wrapper around netfs_io_subrequest

Netfslib has a facility whereby the allocation for netfs_io_subrequest can be increased to so that filesystem-specific data can be tagged on the end.

Prepare to use this by making a struct, cifs_io_subrequest, that wraps netfs_io_subrequest, and absorb struct cifs_readdata into it.

---
 fs/smb/client/cifsglob.h  | 22 ++++++++++---------
 fs/smb/client/cifsproto.h | 13 +++++++++--
 fs/smb/client/cifssmb.c   | 11 ++++------
 fs/smb/client/file.c      | 46 ++++++++++++++++++---------------------
 fs/smb/client/smb2ops.c   |  2 +-
 fs/smb/client/smb2pdu.c   | 12 +++++-----
 fs/smb/client/smb2proto.h |  2 +-
 fs/smb/client/transport.c |  4 ++--
 8 files changed, 59 insertions(+), 53 deletions(-)
fs/smb/client/cifsglob.h +12 -10
index decf80131bbe..1ffd1f3774f7 100644
@@ -256,7 +256,7 @@ struct dfs_info3_param;
 struct cifs_fattr;
 struct smb3_fs_context;
 struct cifs_fid;
-struct cifs_readdata;
+struct cifs_io_subrequest;
 struct cifs_writedata;
 struct cifs_io_parms;
 struct cifs_search_info;
@@ -434,7 +434,7 @@ struct smb_version_operations {
 	/* send a flush request to the server */
 	int (*flush)(const unsigned int, struct cifs_tcon *, struct cifs_fid *);
 	/* async read from the server */
-	int (*async_readv)(struct cifs_readdata *);
+	int (*async_readv)(struct cifs_io_subrequest *);
 	/* async write to the server */
 	int (*async_writev)(struct cifs_writedata *,
 			    void (*release)(struct kref *));
@@ -1465,26 +1465,28 @@ struct cifs_aio_ctx {
 };
 
 /* asynchronous read support */
-struct cifs_readdata {
-	struct kref			refcount;
-	struct list_head		list;
-	struct completion		done;
+struct cifs_io_subrequest {
+	struct netfs_io_subrequest	subreq;
 	struct cifsFileInfo		*cfile;
 	struct address_space		*mapping;
 	struct cifs_aio_ctx		*ctx;
-	__u64				offset;
 	ssize_t				got_bytes;
-	unsigned int			bytes;
 	pid_t				pid;
 	int				result;
-	struct work_struct		work;
-	struct iov_iter			iter;
 	struct kvec			iov[2];
 	struct TCP_Server_Info		*server;
 #ifdef CONFIG_CIFS_SMB_DIRECT
 	struct smbd_mr			*mr;
 #endif
 	struct cifs_credits		credits;
+
+	// TODO: Remove following elements
+	struct list_head		list;
+	struct completion		done;
+	struct work_struct		work;
+	struct iov_iter			iter;
+	__u64				offset;
+	unsigned int			bytes;
 };
 
 /* asynchronous write support */
fs/smb/client/cifsproto.h +11 -2
index 770db9026850..adf4a1f523e3 100644
@@ -590,8 +590,17 @@ void __cifs_put_smb_ses(struct cifs_ses *ses);
 extern struct cifs_ses *
 cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx);
 
-void cifs_readdata_release(struct kref *refcount);
-int cifs_async_readv(struct cifs_readdata *rdata);
+void cifs_readdata_release(struct cifs_io_subrequest *rdata);
+static inline void cifs_get_readdata(struct cifs_io_subrequest *rdata)
+{
+	refcount_inc(&rdata->subreq.ref);
+}
+static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata)
+{
+	if (refcount_dec_and_test(&rdata->subreq.ref))
+		cifs_readdata_release(rdata);
+}
+int cifs_async_readv(struct cifs_io_subrequest *rdata);
 int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid);
 
 int cifs_async_writev(struct cifs_writedata *wdata,
fs/smb/client/cifssmb.c +4 -7
index 01e89070df5a..2d10c07c1295 100644
@@ -24,6 +24,8 @@
 #include <linux/swap.h>
 #include <linux/task_io_accounting_ops.h>
 #include <linux/uaccess.h>
+#include <linux/netfs.h>
+#include <trace/events/netfs.h>
 #include "cifspdu.h"
 #include "cifsfs.h"
 #include "cifsglob.h"
@@ -1262,12 +1264,11 @@ CIFS_open(const unsigned int xid, struct cifs_open_parms *oparms, int *oplock,
 static void
 cifs_readv_callback(struct mid_q_entry *mid)
 {
-	struct cifs_readdata *rdata = mid->callback_data;
+	struct cifs_io_subrequest *rdata = mid->callback_data;
 	struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink);
 	struct TCP_Server_Info *server = tcon->ses->server;
 	struct smb_rqst rqst = { .rq_iov = rdata->iov,
 				 .rq_nvec = 2,
-				 .rq_iter_size = iov_iter_count(&rdata->iter),
 				 .rq_iter = rdata->iter };
 	struct cifs_credits credits = { .value = 1, .instance = 0 };
 
@@ -1312,7 +1313,7 @@ cifs_readv_callback(struct mid_q_entry *mid)
 
 /* cifs_async_readv - send an async write, and set up mid to handle result */
 int
-cifs_async_readv(struct cifs_readdata *rdata)
+cifs_async_readv(struct cifs_io_subrequest *rdata)
 {
 	int rc;
 	READ_REQ *smb = NULL;
@@ -1364,15 +1365,11 @@ cifs_async_readv(struct cifs_readdata *rdata)
 	rdata->iov[1].iov_base = (char *)smb + 4;
 	rdata->iov[1].iov_len = get_rfc1002_length(smb);
 
-	kref_get(&rdata->refcount);
 	rc = cifs_call_async(tcon->ses->server, &rqst, cifs_readv_receive,
 			     cifs_readv_callback, NULL, rdata, 0, NULL);
 
 	if (rc == 0)
 		cifs_stats_inc(&tcon->stats.cifs_stats.num_reads);
-	else
-		kref_put(&rdata->refcount, cifs_readdata_release);
-
 	cifs_small_buf_release(smb);
 	return rc;
 }
fs/smb/client/file.c +21 -25

144 lines not on this page · raw message

fs/smb/client/smb2ops.c +1 -1
index 708973ec750b..9da9ee98ae39 100644
@@ -4478,7 +4478,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid,
 	unsigned int cur_off;
 	unsigned int cur_page_idx;
 	unsigned int pad_len;
-	struct cifs_readdata *rdata = mid->callback_data;
+	struct cifs_io_subrequest *rdata = mid->callback_data;
 	struct smb2_hdr *shdr = (struct smb2_hdr *)buf;
 	int length;
 	bool use_rdma_mr = false;
fs/smb/client/smb2pdu.c +7 -5
index cad402b426b7..3301a80a0f7e 100644
@@ -23,6 +23,8 @@
 #include <linux/uuid.h>
 #include <linux/pagemap.h>
 #include <linux/xattr.h>
+#include <linux/netfs.h>
+#include <trace/events/netfs.h>
 #include "cifsglob.h"
 #include "cifsacl.h"
 #include "cifsproto.h"
@@ -4356,7 +4358,7 @@ static inline bool smb3_use_rdma_offload(struct cifs_io_parms *io_parms)
  */
 static int
 smb2_new_read_req(void **buf, unsigned int *total_len,
-	struct cifs_io_parms *io_parms, struct cifs_readdata *rdata,
+	struct cifs_io_parms *io_parms, struct cifs_io_subrequest *rdata,
 	unsigned int remaining_bytes, int request_type)
 {
 	int rc = -EACCES;
@@ -4448,7 +4450,7 @@ smb2_new_read_req(void **buf, unsigned int *total_len,
 static void
 smb2_readv_callback(struct mid_q_entry *mid)
 {
-	struct cifs_readdata *rdata = mid->callback_data;
+	struct cifs_io_subrequest *rdata = mid->callback_data;
 	struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink);
 	struct TCP_Server_Info *server = rdata->server;
 	struct smb2_hdr *shdr =
@@ -4535,7 +4537,7 @@ smb2_readv_callback(struct mid_q_entry *mid)
 
 /* smb2_async_readv - send an async read, and set up mid to handle result */
 int
-smb2_async_readv(struct cifs_readdata *rdata)
+smb2_async_readv(struct cifs_io_subrequest *rdata)
 {
 	int rc, flags = 0;
 	char *buf;
@@ -4593,13 +4595,13 @@ smb2_async_readv(struct cifs_readdata *rdata)
 		flags |= CIFS_HAS_CREDITS;
 	}
 
-	kref_get(&rdata->refcount);
+	cifs_get_readdata(rdata);
 	rc = cifs_call_async(server, &rqst,
 			     cifs_readv_receive, smb2_readv_callback,
 			     smb3_handle_read_data, rdata, flags,
 			     &rdata->credits);
 	if (rc) {
-		kref_put(&rdata->refcount, cifs_readdata_release);
+		cifs_put_readdata(rdata);
 		cifs_stats_fail_inc(io_parms.tcon, SMB2_READ_HE);
 		trace_smb3_read_err(0 /* xid */, io_parms.persistent_fid,
 				    io_parms.tcon->tid,
fs/smb/client/smb2proto.h +1 -1
index d910d38671ee..6df87d294fbd 100644
@@ -208,7 +208,7 @@ extern int SMB2_query_acl(const unsigned int xid, struct cifs_tcon *tcon,
 extern int SMB2_get_srv_num(const unsigned int xid, struct cifs_tcon *tcon,
 			    u64 persistent_fid, u64 volatile_fid,
 			    __le64 *uniqueid);
-extern int smb2_async_readv(struct cifs_readdata *rdata);
+extern int smb2_async_readv(struct cifs_io_subrequest *rdata);
 extern int SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms,
 		     unsigned int *nbytes, char **buf, int *buf_type);
 extern int smb2_async_writev(struct cifs_writedata *wdata,
fs/smb/client/transport.c +2 -2
index 994d70193432..a5bab478e6de 100644
@@ -1687,7 +1687,7 @@ __cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid,
 static int
 cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 {
-	struct cifs_readdata *rdata = mid->callback_data;
+	struct cifs_io_subrequest *rdata = mid->callback_data;
 
 	return  __cifs_readv_discard(server, mid, rdata->result);
 }
@@ -1697,7 +1697,7 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 {
 	int length, len;
 	unsigned int data_offset, data_len;
-	struct cifs_readdata *rdata = mid->callback_data;
+	struct cifs_io_subrequest *rdata = mid->callback_data;
 	char *buf = server->smallbuf;
 	unsigned int buflen = server->pdu_size + HEADER_PREAMBLE_SIZE(server);
 	bool use_rdma_mr = false;
author maintainer
#2 lore raw

PATCH v5 2/12 cifs: Set zero_point in the copy_file_range() and remap_file_range()

In reply to David Howells

Set zero_point in the copy_file_range() and remap_file_range() implementations so that we don't skip reading data modified on a server-side copy.

---
 fs/smb/client/cifsfs.c | 6 ++++++
 1 file changed, 6 insertions(+)
fs/smb/client/cifsfs.c +6 -0
index 2a4a4e3a8751..41617541d175 100644
@@ -1340,6 +1340,8 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off,
 	rc = cifs_flush_folio(target_inode, destend, &fstart, &fend, false);
 	if (rc)
 		goto unlock;
+	if (fend > target_cifsi->netfs.zero_point)
+		target_cifsi->netfs.zero_point = fend + 1;
 
 	/* Discard all the folios that overlap the destination region. */
 	cifs_dbg(FYI, "about to discard pages %llx-%llx\n", fstart, fend);
@@ -1358,6 +1360,8 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off,
 			fscache_resize_cookie(cifs_inode_cookie(target_inode),
 					      new_size);
 		}
+		if (rc == 0 && new_size > target_cifsi->netfs.zero_point)
+			target_cifsi->netfs.zero_point = new_size;
 	}
 
 	/* force revalidate of size and timestamps of target file now
@@ -1449,6 +1453,8 @@ ssize_t cifs_file_copychunk_range(unsigned int xid,
 	rc = cifs_flush_folio(target_inode, destend, &fstart, &fend, false);
 	if (rc)
 		goto unlock;
+	if (fend > target_cifsi->netfs.zero_point)
+		target_cifsi->netfs.zero_point = fend + 1;
 
 	/* Discard all the folios that overlap the destination region. */
 	truncate_inode_pages_range(&target_inode->i_data, fstart, fend);
author maintainer
#3 lore raw

PATCH v5 3/12 cifs: Replace cifs_writedata with a wrapper around netfs_io_subrequest

In reply to David Howells

Replace the cifs_writedata struct with the same wrapper around netfs_io_subrequest that was used to replace cifs_readdata.

---
 fs/smb/client/cifsglob.h  | 32 +++-------------
 fs/smb/client/cifsproto.h | 16 ++++++--
 fs/smb/client/cifssmb.c   |  9 ++---
 fs/smb/client/file.c      | 79 ++++++++++++++++-----------------------
 fs/smb/client/smb2pdu.c   |  9 ++---
 fs/smb/client/smb2proto.h |  3 +-
 6 files changed, 59 insertions(+), 89 deletions(-)
fs/smb/client/cifsglob.h +6 -26
index 1ffd1f3774f7..7f7073813612 100644
@@ -257,7 +257,6 @@ struct cifs_fattr;
 struct smb3_fs_context;
 struct cifs_fid;
 struct cifs_io_subrequest;
-struct cifs_writedata;
 struct cifs_io_parms;
 struct cifs_search_info;
 struct cifsInodeInfo;
@@ -436,8 +435,7 @@ struct smb_version_operations {
 	/* async read from the server */
 	int (*async_readv)(struct cifs_io_subrequest *);
 	/* async write to the server */
-	int (*async_writev)(struct cifs_writedata *,
-			    void (*release)(struct kref *));
+	int (*async_writev)(struct cifs_io_subrequest *);
 	/* sync read from the server */
 	int (*sync_read)(const unsigned int, struct cifs_fid *,
 			 struct cifs_io_parms *, unsigned int *, char **,
@@ -1480,36 +1478,18 @@ struct cifs_io_subrequest {
 #endif
 	struct cifs_credits		credits;
 
-	// TODO: Remove following elements
-	struct list_head		list;
-	struct completion		done;
-	struct work_struct		work;
-	struct iov_iter			iter;
-	__u64				offset;
-	unsigned int			bytes;
-};
+	enum writeback_sync_modes	sync_mode;
+	bool				uncached;
+	bool				replay;
+	struct bio_vec			*bv;
 
-/* asynchronous write support */
-struct cifs_writedata {
-	struct kref			refcount;
+	// TODO: Remove following elements
 	struct list_head		list;
 	struct completion		done;
-	enum writeback_sync_modes	sync_mode;
 	struct work_struct		work;
-	struct cifsFileInfo		*cfile;
-	struct cifs_aio_ctx		*ctx;
 	struct iov_iter			iter;
-	struct bio_vec			*bv;
 	__u64				offset;
-	pid_t				pid;
 	unsigned int			bytes;
-	int				result;
-	struct TCP_Server_Info		*server;
-#ifdef CONFIG_CIFS_SMB_DIRECT
-	struct smbd_mr			*mr;
-#endif
-	struct cifs_credits		credits;
-	bool				replay;
 };
 
 /*
fs/smb/client/cifsproto.h +12 -4
index adf4a1f523e3..867077be859c 100644
@@ -603,11 +603,19 @@ static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata)
 int cifs_async_readv(struct cifs_io_subrequest *rdata);
 int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid);
 
-int cifs_async_writev(struct cifs_writedata *wdata,
-		      void (*release)(struct kref *kref));
+int cifs_async_writev(struct cifs_io_subrequest *wdata);
 void cifs_writev_complete(struct work_struct *work);
-struct cifs_writedata *cifs_writedata_alloc(work_func_t complete);
-void cifs_writedata_release(struct kref *refcount);
+struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete);
+void cifs_writedata_release(struct cifs_io_subrequest *rdata);
+static inline void cifs_get_writedata(struct cifs_io_subrequest *wdata)
+{
+	refcount_inc(&wdata->subreq.ref);
+}
+static inline void cifs_put_writedata(struct cifs_io_subrequest *wdata)
+{
+	if (refcount_dec_and_test(&wdata->subreq.ref))
+		cifs_writedata_release(wdata);
+}
 int cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 			  struct cifs_sb_info *cifs_sb,
 			  const unsigned char *path, char *pbuf,
fs/smb/client/cifssmb.c +4 -5
index 2d10c07c1295..1ead867a0f4a 100644
@@ -1612,7 +1612,7 @@ CIFSSMBWrite(const unsigned int xid, struct cifs_io_parms *io_parms,
 static void
 cifs_writev_callback(struct mid_q_entry *mid)
 {
-	struct cifs_writedata *wdata = mid->callback_data;
+	struct cifs_io_subrequest *wdata = mid->callback_data;
 	struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink);
 	unsigned int written;
 	WRITE_RSP *smb = (WRITE_RSP *)mid->resp_buf;
@@ -1657,8 +1657,7 @@ cifs_writev_callback(struct mid_q_entry *mid)
 
 /* cifs_async_writev - send an async write, and set up mid to handle result */
 int
-cifs_async_writev(struct cifs_writedata *wdata,
-		  void (*release)(struct kref *kref))
+cifs_async_writev(struct cifs_io_subrequest *wdata)
 {
 	int rc = -EACCES;
 	WRITE_REQ *smb = NULL;
@@ -1725,14 +1724,14 @@ cifs_async_writev(struct cifs_writedata *wdata,
 		iov[1].iov_len += 4; /* pad bigger by four bytes */
 	}
 
-	kref_get(&wdata->refcount);
+	cifs_get_writedata(wdata);
 	rc = cifs_call_async(tcon->ses->server, &rqst, NULL,
 			     cifs_writev_callback, NULL, wdata, 0, NULL);
 
 	if (rc == 0)
 		cifs_stats_inc(&tcon->stats.cifs_stats.num_writes);
 	else
-		kref_put(&wdata->refcount, release);
+		cifs_put_writedata(wdata);
 
 async_writev_out:
 	cifs_small_buf_release(smb);
fs/smb/client/file.c +32 -47

226 lines not on this page · raw message

fs/smb/client/smb2pdu.c +4 -5
index 3301a80a0f7e..db739806343d 100644
@@ -4702,7 +4702,7 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms,
 static void
 smb2_writev_callback(struct mid_q_entry *mid)
 {
-	struct cifs_writedata *wdata = mid->callback_data;
+	struct cifs_io_subrequest *wdata = mid->callback_data;
 	struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink);
 	struct TCP_Server_Info *server = wdata->server;
 	unsigned int written;
@@ -4783,8 +4783,7 @@ smb2_writev_callback(struct mid_q_entry *mid)
 
 /* smb2_async_writev - send an async write, and set up mid to handle result */
 int
-smb2_async_writev(struct cifs_writedata *wdata,
-		  void (*release)(struct kref *kref))
+smb2_async_writev(struct cifs_io_subrequest *wdata)
 {
 	int rc = -EACCES, flags = 0;
 	struct smb2_write_req *req = NULL;
@@ -4918,7 +4917,7 @@ smb2_async_writev(struct cifs_writedata *wdata,
 		flags |= CIFS_HAS_CREDITS;
 	}
 
-	kref_get(&wdata->refcount);
+	cifs_get_writedata(wdata);
 	rc = cifs_call_async(server, &rqst, NULL, smb2_writev_callback, NULL,
 			     wdata, flags, &wdata->credits);
 
@@ -4930,7 +4929,7 @@ smb2_async_writev(struct cifs_writedata *wdata,
 				     io_parms->offset,
 				     io_parms->length,
 				     rc);
-		kref_put(&wdata->refcount, release);
+		cifs_put_writedata(wdata);
 		cifs_stats_fail_inc(tcon, SMB2_WRITE_HE);
 	}
 
fs/smb/client/smb2proto.h +1 -2
index 6df87d294fbd..39ac8ed06281 100644
@@ -211,8 +211,7 @@ extern int SMB2_get_srv_num(const unsigned int xid, struct cifs_tcon *tcon,
 extern int smb2_async_readv(struct cifs_io_subrequest *rdata);
 extern int SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms,
 		     unsigned int *nbytes, char **buf, int *buf_type);
-extern int smb2_async_writev(struct cifs_writedata *wdata,
-			     void (*release)(struct kref *kref));
+extern int smb2_async_writev(struct cifs_io_subrequest *wdata);
 extern int SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms,
 		      unsigned int *nbytes, struct kvec *iov, int n_vec);
 extern int SMB2_echo(struct TCP_Server_Info *server);
author maintainer
#4 lore raw

PATCH v5 4/12 cifs: Use more fields from netfs_io_subrequest

In reply to David Howells

Use more fields from netfs_io_subrequest instead of those incorporated into cifs_io_subrequest from cifs_readdata and cifs_writedata.

---
 fs/smb/client/cifsglob.h  |   3 -
 fs/smb/client/cifssmb.c   |  52 +++++++++---------
 fs/smb/client/file.c      | 112 +++++++++++++++++++-------------------
 fs/smb/client/smb2ops.c   |   4 +-
 fs/smb/client/smb2pdu.c   |  52 +++++++++---------
 fs/smb/client/transport.c |   6 +-
 6 files changed, 113 insertions(+), 116 deletions(-)
fs/smb/client/cifsglob.h +0 -3
index 7f7073813612..cac10f8e17e4 100644
@@ -1487,9 +1487,6 @@ struct cifs_io_subrequest {
 	struct list_head		list;
 	struct completion		done;
 	struct work_struct		work;
-	struct iov_iter			iter;
-	__u64				offset;
-	unsigned int			bytes;
 };
 
 /*
fs/smb/client/cifssmb.c +26 -26

117 lines not on this page · raw message

fs/smb/client/file.c +56 -56

284 lines not on this page · raw message

fs/smb/client/smb2ops.c +2 -2
index 9da9ee98ae39..8d674aef8dd9 100644
@@ -4580,7 +4580,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid,
 
 		/* Copy the data to the output I/O iterator. */
 		rdata->result = cifs_copy_pages_to_iter(pages, pages_len,
-							cur_off, &rdata->iter);
+							cur_off, &rdata->subreq.io_iter);
 		if (rdata->result != 0) {
 			if (is_offloaded)
 				mid->mid_state = MID_RESPONSE_MALFORMED;
@@ -4594,7 +4594,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid,
 		/* read response payload is in buf */
 		WARN_ONCE(pages && !xa_empty(pages),
 			  "read data can be either in buf or in pages");
-		length = copy_to_iter(buf + data_offset, data_len, &rdata->iter);
+		length = copy_to_iter(buf + data_offset, data_len, &rdata->subreq.io_iter);
 		if (length < 0)
 			return length;
 		rdata->got_bytes = data_len;
fs/smb/client/smb2pdu.c +26 -26

153 lines not on this page · raw message

fs/smb/client/transport.c +3 -3
index a5bab478e6de..e52967de59e6 100644
@@ -1702,8 +1702,8 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 	unsigned int buflen = server->pdu_size + HEADER_PREAMBLE_SIZE(server);
 	bool use_rdma_mr = false;
 
-	cifs_dbg(FYI, "%s: mid=%llu offset=%llu bytes=%u\n",
-		 __func__, mid->mid, rdata->offset, rdata->bytes);
+	cifs_dbg(FYI, "%s: mid=%llu offset=%llu bytes=%zu\n",
+		 __func__, mid->mid, rdata->subreq.start, rdata->subreq.len);
 
 	/*
 	 * read the rest of READ_RSP header (sans Data array), or whatever we
@@ -1808,7 +1808,7 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 		length = data_len; /* An RDMA read is already done. */
 	else
 #endif
-		length = cifs_read_iter_from_socket(server, &rdata->iter,
+		length = cifs_read_iter_from_socket(server, &rdata->subreq.io_iter,
 						    data_len);
 	if (length > 0)
 		rdata->got_bytes += length;
author maintainer
#5 lore raw

PATCH v5 5/12 cifs: Make wait_mtu_credits take size_t args

In reply to David Howells

Make the wait_mtu_credits functions use size_t for the size and num arguments rather than unsigned int as netfslib uses size_t/ssize_t for arguments and return values to allow for extra capacity.

---
 fs/smb/client/cifsglob.h  |  4 ++--
 fs/smb/client/cifsproto.h |  2 +-
 fs/smb/client/file.c      | 18 ++++++++++--------
 fs/smb/client/smb2ops.c   |  4 ++--
 fs/smb/client/transport.c |  4 ++--
 5 files changed, 17 insertions(+), 15 deletions(-)
fs/smb/client/cifsglob.h +2 -2
index cac10f8e17e4..c0a0de24f52a 100644
@@ -530,8 +530,8 @@ struct smb_version_operations {
 	/* writepages retry size */
 	unsigned int (*wp_retry_size)(struct inode *);
 	/* get mtu credits */
-	int (*wait_mtu_credits)(struct TCP_Server_Info *, unsigned int,
-				unsigned int *, struct cifs_credits *);
+	int (*wait_mtu_credits)(struct TCP_Server_Info *, size_t,
+				size_t *, struct cifs_credits *);
 	/* adjust previously taken mtu credits to request size */
 	int (*adjust_credits)(struct TCP_Server_Info *server,
 			      struct cifs_credits *credits,
fs/smb/client/cifsproto.h +1 -1
index 867077be859c..00cb0d2dc935 100644
@@ -121,7 +121,7 @@ extern struct mid_q_entry *cifs_setup_async_request(struct TCP_Server_Info *,
 extern int cifs_check_receive(struct mid_q_entry *mid,
 			struct TCP_Server_Info *server, bool log_error);
 extern int cifs_wait_mtu_credits(struct TCP_Server_Info *server,
-				 unsigned int size, unsigned int *num,
+				 size_t size, size_t *num,
 				 struct cifs_credits *credits);
 extern int SendReceive2(const unsigned int /* xid */ , struct cifs_ses *,
 			struct kvec *, int /* nvec to send */,
fs/smb/client/file.c +10 -8
index 6e53657154d6..63237dad0c4d 100644
@@ -2735,9 +2735,9 @@ 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;
+	unsigned int xid, len;
 	loff_t i_size = i_size_read(inode);
-	size_t max_len;
+	size_t max_len, wsize;
 	long count = wbc->nr_to_write;
 	int rc;
 
@@ -3249,7 +3249,7 @@ static int
 cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list,
 	struct cifs_aio_ctx *ctx)
 {
-	unsigned int wsize;
+	size_t wsize;
 	struct cifs_credits credits;
 	int rc;
 	struct TCP_Server_Info *server = wdata->server;
@@ -3384,7 +3384,8 @@ cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from,
 	do {
 		struct cifs_credits credits_on_stack;
 		struct cifs_credits *credits = &credits_on_stack;
-		unsigned int wsize, nsegs = 0;
+		unsigned int nsegs = 0;
+		size_t wsize;
 
 		if (signal_pending(current)) {
 			rc = -EINTR;
@@ -3821,7 +3822,7 @@ static int cifs_resend_rdata(struct cifs_io_subrequest *rdata,
 			struct list_head *rdata_list,
 			struct cifs_aio_ctx *ctx)
 {
-	unsigned int rsize;
+	size_t rsize;
 	struct cifs_credits credits;
 	int rc;
 	struct TCP_Server_Info *server;
@@ -3895,10 +3896,10 @@ cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file,
 		     struct cifs_aio_ctx *ctx)
 {
 	struct cifs_io_subrequest *rdata;
-	unsigned int rsize, nsegs, max_segs = INT_MAX;
+	unsigned int nsegs, max_segs = INT_MAX;
 	struct cifs_credits credits_on_stack;
 	struct cifs_credits *credits = &credits_on_stack;
-	size_t cur_len, max_len;
+	size_t cur_len, max_len, rsize;
 	int rc;
 	pid_t pid;
 	struct TCP_Server_Info *server;
@@ -4494,12 +4495,13 @@ static void cifs_readahead(struct readahead_control *ractl)
 	 * Chop the readahead request up into rsize-sized read requests.
 	 */
 	while ((nr_pages = ra_pages)) {
-		unsigned int i, rsize;
+		unsigned int i;
 		struct cifs_io_subrequest *rdata;
 		struct cifs_credits credits_on_stack;
 		struct cifs_credits *credits = &credits_on_stack;
 		struct folio *folio;
 		pgoff_t fsize;
+		size_t rsize;
 
 		/*
 		 * Find out if we have anything cached in the range of
fs/smb/client/smb2ops.c +2 -2
index 8d674aef8dd9..b79b6bbd7103 100644
@@ -217,8 +217,8 @@ smb2_get_credits(struct mid_q_entry *mid)
 }
 
 static int
-smb2_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size,
-		      unsigned int *num, struct cifs_credits *credits)
+smb2_wait_mtu_credits(struct TCP_Server_Info *server, size_t size,
+		      size_t *num, struct cifs_credits *credits)
 {
 	int rc = 0;
 	unsigned int scredits, in_flight;
fs/smb/client/transport.c +2 -2
index e52967de59e6..5a69a7430ffa 100644
@@ -691,8 +691,8 @@ wait_for_compound_request(struct TCP_Server_Info *server, int num,
 }
 
 int
-cifs_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size,
-		      unsigned int *num, struct cifs_credits *credits)
+cifs_wait_mtu_credits(struct TCP_Server_Info *server, size_t size,
+		      size_t *num, struct cifs_credits *credits)
 {
 	*num = size;
 	credits->value = 0;
author maintainer
#6 lore raw

PATCH v5 6/12 cifs: Implement netfslib hooks

In reply to David Howells

Provide implementation of the netfslib hooks that will be used by netfslib to ask cifs to set up and perform operations. Of particular note are

(*) cifs_clamp_length() - This is used to negotiate the size of the next
subrequest in a read request, taking into account the credit available and the rsize. The credits are attached to the subrequest.

(*) cifs_req_issue_read() - This is used to issue a subrequest that has
been set up and clamped.

(*) cifs_create_write_requests() - This is used to break the given span of
file positions into suboperations according to cifs's wsize and available credits. As each subop is created, it can be dispatched or queued for dispatch.

At this point, cifs is not wired up to actually *use* netfslib; that will be done in a subsequent patch.

---
 fs/netfs/buffered_write.c    |   3 +
 fs/smb/client/Kconfig        |   1 +
 fs/smb/client/cifsglob.h     |  28 ++-
 fs/smb/client/file.c         | 357 +++++++++++++++++++++++++++++++++++
 include/linux/netfs.h        |   1 +
 include/trace/events/netfs.h |   1 +
 6 files changed, 382 insertions(+), 9 deletions(-)
fs/netfs/buffered_write.c +3 -0
index a3059b3168fd..2c117b825247 100644
@@ -394,6 +394,9 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter,
 	} while (iov_iter_count(iter));
 
 out:
+	if (likely(written) && ctx->ops->post_modify)
+		ctx->ops->post_modify(inode);
+
 	if (unlikely(wreq)) {
 		ret = netfs_end_writethrough(wreq, iocb);
 		wbc_detach_inode(&wbc);
fs/smb/client/Kconfig +1 -0
index 2927bd174a88..2517dc242386 100644
@@ -2,6 +2,7 @@
 config CIFS
 	tristate "SMB3 and CIFS support (advanced network filesystem)"
 	depends on INET
+	select NETFS_SUPPORT
 	select NLS
 	select NLS_UCS2_UTILS
 	select CRYPTO
fs/smb/client/cifsglob.h +19 -9
index c0a0de24f52a..dd0cef742a64 100644
@@ -1462,15 +1462,24 @@ struct cifs_aio_ctx {
 	bool			direct_io;
 };
 
+struct cifs_io_request {
+	struct netfs_io_request		rreq;
+	struct cifsFileInfo		*cfile;
+};
+
 /* asynchronous read support */
 struct cifs_io_subrequest {
-	struct netfs_io_subrequest	subreq;
-	struct cifsFileInfo		*cfile;
-	struct address_space		*mapping;
-	struct cifs_aio_ctx		*ctx;
+	union {
+		struct netfs_io_subrequest subreq;
+		struct netfs_io_request *rreq;
+		struct cifs_io_request *req;
+	};
 	ssize_t				got_bytes;
 	pid_t				pid;
+	unsigned int			xid;
 	int				result;
+	bool				have_credits;
+	bool				replay;
 	struct kvec			iov[2];
 	struct TCP_Server_Info		*server;
 #ifdef CONFIG_CIFS_SMB_DIRECT
@@ -1478,15 +1487,16 @@ struct cifs_io_subrequest {
 #endif
 	struct cifs_credits		credits;
 
-	enum writeback_sync_modes	sync_mode;
-	bool				uncached;
-	bool				replay;
-	struct bio_vec			*bv;
-
 	// TODO: Remove following elements
 	struct list_head		list;
 	struct completion		done;
 	struct work_struct		work;
+	struct cifsFileInfo		*cfile;
+	struct address_space		*mapping;
+	struct cifs_aio_ctx		*ctx;
+	enum writeback_sync_modes	sync_mode;
+	bool				uncached;
+	struct bio_vec			*bv;
 };
 
 /*
fs/smb/client/file.c +357 -0

363 lines not on this page · raw message

include/linux/netfs.h +1 -0
index 100cbb261269..61195cf16d6e 100644
@@ -312,6 +312,7 @@ struct netfs_request_ops {
 
 	/* Modification handling */
 	void (*update_i_size)(struct inode *inode, loff_t i_size);
+	void (*post_modify)(struct inode *inode);
 
 	/* Write request handling */
 	void (*create_write_requests)(struct netfs_io_request *wreq,
include/trace/events/netfs.h +1 -0
index 447a8c21cf57..06567b5be8fa 100644
@@ -101,6 +101,7 @@
 #define netfs_sreq_ref_traces					\
 	EM(netfs_sreq_trace_get_copy_to_cache,	"GET COPY2C ")	\
 	EM(netfs_sreq_trace_get_resubmit,	"GET RESUBMIT")	\
+	EM(netfs_sreq_trace_get_submit,		"GET SUBMIT")	\
 	EM(netfs_sreq_trace_get_short_read,	"GET SHORTRD")	\
 	EM(netfs_sreq_trace_new,		"NEW        ")	\
 	EM(netfs_sreq_trace_put_clear,		"PUT CLEAR  ")	\
author maintainer
#7 lore raw

PATCH v5 7/12 cifs: Replace the writedata replay bool with a netfs sreq flag

In reply to David Howells

Replace the 'replay' bool in cifs_writedata (now cifs_io_subrequest) with a flag in the netfs_io_subrequest flags.

---
 fs/smb/client/cifsglob.h | 1 -
 fs/smb/client/file.c     | 2 +-
 fs/smb/client/smb2pdu.c  | 4 ++--
 include/linux/netfs.h    | 1 +
 4 files changed, 4 insertions(+), 4 deletions(-)
fs/smb/client/cifsglob.h +0 -1
index dd0cef742a64..1dfed3eddaa2 100644
@@ -1479,7 +1479,6 @@ struct cifs_io_subrequest {
 	unsigned int			xid;
 	int				result;
 	bool				have_credits;
-	bool				replay;
 	struct kvec			iov[2];
 	struct TCP_Server_Info		*server;
 #ifdef CONFIG_CIFS_SMB_DIRECT
fs/smb/client/file.c +1 -1
index ce5f24206be0..14602ed6bc39 100644
@@ -3645,7 +3645,7 @@ cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list
 			if (wdata->cfile->invalidHandle)
 				rc = -EAGAIN;
 			else {
-				wdata->replay = true;
+				set_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags);
 #ifdef CONFIG_CIFS_SMB_DIRECT
 				if (wdata->mr) {
 					wdata->mr->need_invalidate = true;
fs/smb/client/smb2pdu.c +2 -2
index 2ecc5f210329..84e3675eb41e 100644
@@ -4797,7 +4797,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata)
 	struct cifs_io_parms *io_parms = NULL;
 	int credit_request;
 
-	if (!wdata->server || wdata->replay)
+	if (!wdata->server || test_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags))
 		server = wdata->server = cifs_pick_channel(tcon->ses);
 
 	/*
@@ -4882,7 +4882,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata)
 	rqst.rq_nvec = 1;
 	rqst.rq_iter = wdata->subreq.io_iter;
 	rqst.rq_iter_size = iov_iter_count(&rqst.rq_iter);
-	if (wdata->replay)
+	if (test_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags))
 		smb2_set_replay(server, &rqst);
 #ifdef CONFIG_CIFS_SMB_DIRECT
 	if (wdata->mr)
include/linux/netfs.h +1 -0
index 61195cf16d6e..455ccfe8bffa 100644
@@ -224,6 +224,7 @@ struct netfs_io_subrequest {
 #define NETFS_SREQ_SEEK_DATA_READ	3	/* Set if ->read() should SEEK_DATA first */
 #define NETFS_SREQ_NO_PROGRESS		4	/* Set if we didn't manage to read any data */
 #define NETFS_SREQ_ONDEMAND		5	/* Set if it's from on-demand read mode */
+#define NETFS_SREQ_RETRYING		6	/* Set if we're retrying the op */
 };
 
 enum netfs_io_origin {
author maintainer
#8 lore raw

PATCH v5 8/12 cifs: Move cifs_loose_read_iter() and cifs_file_write_iter() to file.c

In reply to David Howells

Move cifs_loose_read_iter() and cifs_file_write_iter() to file.c so that they are colocated with similar functions rather than being split with cifsfs.c.

---
 fs/smb/client/cifsfs.c | 55 ------------------------------------------
 fs/smb/client/cifsfs.h |  2 ++
 fs/smb/client/file.c   | 53 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 55 insertions(+), 55 deletions(-)
fs/smb/client/cifsfs.c +0 -55
index 41617541d175..4b6d1a5e4741 100644
@@ -984,61 +984,6 @@ cifs_smb3_do_mount(struct file_system_type *fs_type,
 	return root;
 }
 
-
-static ssize_t
-cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter)
-{
-	ssize_t rc;
-	struct inode *inode = file_inode(iocb->ki_filp);
-
-	if (iocb->ki_flags & IOCB_DIRECT)
-		return cifs_user_readv(iocb, iter);
-
-	rc = cifs_revalidate_mapping(inode);
-	if (rc)
-		return rc;
-
-	return generic_file_read_iter(iocb, iter);
-}
-
-static ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
-{
-	struct inode *inode = file_inode(iocb->ki_filp);
-	struct cifsInodeInfo *cinode = CIFS_I(inode);
-	ssize_t written;
-	int rc;
-
-	if (iocb->ki_filp->f_flags & O_DIRECT) {
-		written = cifs_user_writev(iocb, from);
-		if (written > 0 && CIFS_CACHE_READ(cinode)) {
-			cifs_zap_mapping(inode);
-			cifs_dbg(FYI,
-				 "Set no oplock for inode=%p after a write operation\n",
-				 inode);
-			cinode->oplock = 0;
-		}
-		return written;
-	}
-
-	written = cifs_get_writer(cinode);
-	if (written)
-		return written;
-
-	written = generic_file_write_iter(iocb, from);
-
-	if (CIFS_CACHE_WRITE(CIFS_I(inode)))
-		goto out;
-
-	rc = filemap_fdatawrite(inode->i_mapping);
-	if (rc)
-		cifs_dbg(FYI, "cifs_file_write_iter: %d rc on %p inode\n",
-			 rc, inode);
-
-out:
-	cifs_put_writer(cinode);
-	return written;
-}
-
 static loff_t cifs_llseek(struct file *file, loff_t offset, int whence)
 {
 	struct cifsFileInfo *cfile = file->private_data;
fs/smb/client/cifsfs.h +2 -0
index 685f7d1139c6..e8e0f863e935 100644
@@ -100,6 +100,8 @@ extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to);
 extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from);
 extern ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from);
 extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from);
+ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from);
+ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter);
 extern int cifs_flock(struct file *pfile, int cmd, struct file_lock *plock);
 extern int cifs_lock(struct file *, int, struct file_lock *);
 extern int cifs_fsync(struct file *, loff_t, loff_t, int);
fs/smb/client/file.c +53 -0

59 lines not on this page · raw message

author maintainer
#9 lore raw

PATCH v5 9/12 cifs: Cut over to using netfslib

In reply to David Howells

Make the cifs filesystem use netfslib to handle reading and writing on behalf of cifs. The changes include:

(1) Various read_iter/write_iter type functions are turned into wrappers around netfslib API functions or are pointed directly at those functions:

cifs_file_direct{,_nobrl}_ops switch to use netfs_unbuffered_read_iter and netfs_unbuffered_write_iter.

Large pieces of code that will be removed are #if'd out and will be removed in subsequent patches.

[?] Why does cifs mark the page dirty in the destination buffer of a DIO
read? Should that happen automatically? Does netfs need to do that?

---
 fs/netfs/io.c             |   7 +-
 fs/smb/client/cifsfs.c    |   8 +-
 fs/smb/client/cifsfs.h    |   8 +-
 fs/smb/client/cifsglob.h  |   3 +-
 fs/smb/client/cifsproto.h |   8 +-
 fs/smb/client/cifssmb.c   |  45 ++++++-----
 fs/smb/client/file.c      | 166 ++++++++++++++++++++------------------
 fs/smb/client/fscache.c   |   2 +
 fs/smb/client/fscache.h   |   4 +
 fs/smb/client/inode.c     |  19 ++++-
 fs/smb/client/smb2pdu.c   | 100 ++++++++++++++---------
 fs/smb/client/trace.h     | 144 ++++++++++++++++++++++++++++-----
 fs/smb/client/transport.c |   3 +
 13 files changed, 347 insertions(+), 170 deletions(-)
fs/netfs/io.c +6 -1
index e8ff1e61ce79..02f202c6209f 100644
@@ -352,8 +352,13 @@ static void netfs_rreq_assess_dio(struct netfs_io_request *rreq)
 	unsigned int i;
 	size_t transferred = 0;
 
-	for (i = 0; i < rreq->direct_bv_count; i++)
+	for (i = 0; i < rreq->direct_bv_count; i++) {
 		flush_dcache_page(rreq->direct_bv[i].bv_page);
+		// TODO: cifs marks pages in the destination buffer
+		// dirty under some circumstances after a read.  Do we
+		// need to do that too?
+		set_page_dirty(rreq->direct_bv[i].bv_page);
+	}
 
 	list_for_each_entry(subreq, &rreq->subrequests, rreq_link) {
 		if (subreq->error || subreq->transferred == 0)
fs/smb/client/cifsfs.c +4 -4
index 4b6d1a5e4741..de838cb82ce1 100644
@@ -1516,8 +1516,8 @@ const struct file_operations cifs_file_strict_ops = {
 };
 
 const struct file_operations cifs_file_direct_ops = {
-	.read_iter = cifs_direct_readv,
-	.write_iter = cifs_direct_writev,
+	.read_iter = netfs_unbuffered_read_iter,
+	.write_iter = netfs_file_write_iter,
 	.open = cifs_open,
 	.release = cifs_close,
 	.lock = cifs_lock,
@@ -1572,8 +1572,8 @@ const struct file_operations cifs_file_strict_nobrl_ops = {
 };
 
 const struct file_operations cifs_file_direct_nobrl_ops = {
-	.read_iter = cifs_direct_readv,
-	.write_iter = cifs_direct_writev,
+	.read_iter = netfs_unbuffered_read_iter,
+	.write_iter = netfs_file_write_iter,
 	.open = cifs_open,
 	.release = cifs_close,
 	.fsync = cifs_fsync,
fs/smb/client/cifsfs.h +1 -7
index e8e0f863e935..5cd547b7b5ea 100644
@@ -85,6 +85,7 @@ extern const struct inode_operations cifs_namespace_inode_operations;
 
 
 /* Functions related to files and directories */
+extern const struct netfs_request_ops cifs_req_ops;
 extern const struct file_operations cifs_file_ops;
 extern const struct file_operations cifs_file_direct_ops; /* if directio mnt */
 extern const struct file_operations cifs_file_strict_ops; /* if strictio mnt */
@@ -94,11 +95,7 @@ extern const struct file_operations cifs_file_strict_nobrl_ops;
 extern int cifs_open(struct inode *inode, struct file *file);
 extern int cifs_close(struct inode *inode, struct file *file);
 extern int cifs_closedir(struct inode *inode, struct file *file);
-extern ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to);
-extern ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to);
 extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to);
-extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from);
-extern ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from);
 extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from);
 ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from);
 ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter);
@@ -112,9 +109,6 @@ extern int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma);
 extern const struct file_operations cifs_dir_ops;
 extern int cifs_dir_open(struct inode *inode, struct file *file);
 extern int cifs_readdir(struct file *file, struct dir_context *ctx);
-extern void cifs_pages_written_back(struct inode *inode, loff_t start, unsigned int len);
-extern void cifs_pages_write_failed(struct inode *inode, loff_t start, unsigned int len);
-extern void cifs_pages_write_redirty(struct inode *inode, loff_t start, unsigned int len);
 
 /* Functions related to dir entries */
 extern const struct dentry_operations cifs_dentry_ops;
fs/smb/client/cifsglob.h +2 -1
index 1dfed3eddaa2..259dc3893e28 100644
@@ -1486,7 +1486,7 @@ struct cifs_io_subrequest {
 #endif
 	struct cifs_credits		credits;
 
-	// TODO: Remove following elements
+#if 0 // TODO: Remove following elements
 	struct list_head		list;
 	struct completion		done;
 	struct work_struct		work;
@@ -1496,6 +1496,7 @@ struct cifs_io_subrequest {
 	enum writeback_sync_modes	sync_mode;
 	bool				uncached;
 	struct bio_vec			*bv;
+#endif
 };
 
 /*
fs/smb/client/cifsproto.h +6 -2
index 00cb0d2dc935..d869918fb5c8 100644
@@ -145,8 +145,8 @@ 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 void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset,
-			    unsigned int bytes_written);
+void cifs_write_subrequest_terminated(struct cifs_io_subrequest *wdata, ssize_t result,
+				      bool was_async);
 extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, int);
 extern int cifs_get_writable_file(struct cifsInodeInfo *cifs_inode,
 				  int flags,
@@ -590,6 +590,7 @@ void __cifs_put_smb_ses(struct cifs_ses *ses);
 extern struct cifs_ses *
 cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx);
 
+#if 0 // TODO Remove
 void cifs_readdata_release(struct cifs_io_subrequest *rdata);
 static inline void cifs_get_readdata(struct cifs_io_subrequest *rdata)
 {
@@ -600,11 +601,13 @@ static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata)
 	if (refcount_dec_and_test(&rdata->subreq.ref))
 		cifs_readdata_release(rdata);
 }
+#endif
 int cifs_async_readv(struct cifs_io_subrequest *rdata);
 int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid);
 
 int cifs_async_writev(struct cifs_io_subrequest *wdata);
 void cifs_writev_complete(struct work_struct *work);
+#if 0 // TODO Remove
 struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete);
 void cifs_writedata_release(struct cifs_io_subrequest *rdata);
 static inline void cifs_get_writedata(struct cifs_io_subrequest *wdata)
@@ -616,6 +619,7 @@ static inline void cifs_put_writedata(struct cifs_io_subrequest *wdata)
 	if (refcount_dec_and_test(&wdata->subreq.ref))
 		cifs_writedata_release(wdata);
 }
+#endif
 int cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 			  struct cifs_sb_info *cifs_sb,
 			  const unsigned char *path, char *pbuf,
fs/smb/client/cifssmb.c +26 -19

127 lines not on this page · raw message

fs/smb/client/file.c +89 -77

381 lines not on this page · raw message

fs/smb/client/fscache.c +2 -0
index c4a3cb736881..228fe57bbde3 100644
@@ -137,6 +137,7 @@ void cifs_fscache_release_inode_cookie(struct inode *inode)
 	}
 }
 
+#if 0 // TODO remove
 /*
  * Fallback page reading interface.
  */
@@ -245,3 +246,4 @@ int __cifs_fscache_query_occupancy(struct inode *inode,
 	fscache_end_operation(&cres);
 	return ret;
 }
+#endif
fs/smb/client/fscache.h +4 -0
index a3d73720914f..c2c05a778a71 100644
@@ -74,6 +74,7 @@ static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags
 			   i_size_read(inode), flags);
 }
 
+#if 0 // TODO remove
 extern int __cifs_fscache_query_occupancy(struct inode *inode,
 					  pgoff_t first, unsigned int nr_pages,
 					  pgoff_t *_data_first,
@@ -108,6 +109,7 @@ static inline void cifs_readahead_to_fscache(struct inode *inode,
 	if (cifs_inode_cookie(inode))
 		__cifs_readahead_to_fscache(inode, pos, len);
 }
+#endif
 
 #else /* CONFIG_CIFS_FSCACHE */
 static inline
@@ -125,6 +127,7 @@ static inline void cifs_fscache_unuse_inode_cookie(struct inode *inode, bool upd
 static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { return NULL; }
 static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) {}
 
+#if 0 // TODO remove
 static inline int cifs_fscache_query_occupancy(struct inode *inode,
 					       pgoff_t first, unsigned int nr_pages,
 					       pgoff_t *_data_first,
@@ -143,6 +146,7 @@ cifs_readpage_from_fscache(struct inode *inode, struct page *page)
 
 static inline
 void cifs_readahead_to_fscache(struct inode *inode, loff_t pos, size_t len) {}
+#endif
 
 #endif /* CONFIG_CIFS_FSCACHE */
 
fs/smb/client/inode.c +18 -1
index 24489e1e238a..7acaf75feabc 100644
@@ -28,14 +28,29 @@
 #include "cached_dir.h"
 #include "reparse.h"
 
+/*
+ * Set parameters for the netfs library
+ */
+static void cifs_set_netfs_context(struct inode *inode)
+{
+	struct cifsInodeInfo *cifs_i = CIFS_I(inode);
+	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
+
+	netfs_inode_init(&cifs_i->netfs, &cifs_req_ops, true);
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO)
+		__set_bit(NETFS_ICTX_WRITETHROUGH, &cifs_i->netfs.flags);
+}
+
 static void cifs_set_ops(struct inode *inode)
 {
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
+	struct netfs_inode *ictx = netfs_inode(inode);
 
 	switch (inode->i_mode & S_IFMT) {
 	case S_IFREG:
 		inode->i_op = &cifs_file_inode_ops;
 		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) {
+			set_bit(NETFS_ICTX_UNBUFFERED, &ictx->flags);
 			if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
 				inode->i_fop = &cifs_file_direct_nobrl_ops;
 			else
@@ -220,8 +235,10 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr)
 
 	if (fattr->cf_flags & CIFS_FATTR_JUNCTION)
 		inode->i_flags |= S_AUTOMOUNT;
-	if (inode->i_state & I_NEW)
+	if (inode->i_state & I_NEW) {
+		cifs_set_netfs_context(inode);
 		cifs_set_ops(inode);
+	}
 	return 0;
 }
 
fs/smb/client/smb2pdu.c +61 -39

249 lines not on this page · raw message

fs/smb/client/trace.h +125 -19

178 lines not on this page · raw message

fs/smb/client/transport.c +3 -0
index 5a69a7430ffa..4bf8b2ff26f5 100644
@@ -1808,8 +1808,11 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 		length = data_len; /* An RDMA read is already done. */
 	else
 #endif
+	{
 		length = cifs_read_iter_from_socket(server, &rdata->subreq.io_iter,
 						    data_len);
+		iov_iter_revert(&rdata->subreq.io_iter, data_len);
+	}
 	if (length > 0)
 		rdata->got_bytes += length;
 	server->total_read += length;
author maintainer
#16 lore raw

PATCH v5 10/12 cifs: Remove some code that's no longer used, part 1

In reply to David Howells

Remove some code that was #if'd out with the netfslib conversion. This is split into parts for file.c as the diff generator otherwise produces a hard to read diff for part of it where a big chunk is cut out.

---
 fs/smb/client/cifsglob.h  |  12 -
 fs/smb/client/cifsproto.h |  25 --
 fs/smb/client/file.c      | 640 --------------------------------------
 fs/smb/client/fscache.c   | 111 -------
 fs/smb/client/fscache.h   |  58 ----
 5 files changed, 846 deletions(-)
fs/smb/client/cifsglob.h +0 -12
index 259dc3893e28..70f205997006 100644
@@ -1485,18 +1485,6 @@ struct cifs_io_subrequest {
 	struct smbd_mr			*mr;
 #endif
 	struct cifs_credits		credits;
-
-#if 0 // TODO: Remove following elements
-	struct list_head		list;
-	struct completion		done;
-	struct work_struct		work;
-	struct cifsFileInfo		*cfile;
-	struct address_space		*mapping;
-	struct cifs_aio_ctx		*ctx;
-	enum writeback_sync_modes	sync_mode;
-	bool				uncached;
-	struct bio_vec			*bv;
-#endif
 };
 
 /*
fs/smb/client/cifsproto.h +0 -25
index d869918fb5c8..1468df1ff47d 100644
@@ -590,36 +590,11 @@ void __cifs_put_smb_ses(struct cifs_ses *ses);
 extern struct cifs_ses *
 cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx);
 
-#if 0 // TODO Remove
-void cifs_readdata_release(struct cifs_io_subrequest *rdata);
-static inline void cifs_get_readdata(struct cifs_io_subrequest *rdata)
-{
-	refcount_inc(&rdata->subreq.ref);
-}
-static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata)
-{
-	if (refcount_dec_and_test(&rdata->subreq.ref))
-		cifs_readdata_release(rdata);
-}
-#endif
 int cifs_async_readv(struct cifs_io_subrequest *rdata);
 int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid);
 
 int cifs_async_writev(struct cifs_io_subrequest *wdata);
 void cifs_writev_complete(struct work_struct *work);
-#if 0 // TODO Remove
-struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete);
-void cifs_writedata_release(struct cifs_io_subrequest *rdata);
-static inline void cifs_get_writedata(struct cifs_io_subrequest *wdata)
-{
-	refcount_inc(&wdata->subreq.ref);
-}
-static inline void cifs_put_writedata(struct cifs_io_subrequest *wdata)
-{
-	if (refcount_dec_and_test(&wdata->subreq.ref))
-		cifs_writedata_release(wdata);
-}
-#endif
 int cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 			  struct cifs_sb_info *cifs_sb,
 			  const unsigned char *path, char *pbuf,
fs/smb/client/file.c +0 -640

670 lines not on this page · raw message

fs/smb/client/fscache.c +0 -111

114 lines not on this page · raw message

fs/smb/client/fscache.h +0 -58

70 lines not on this page · raw message

author maintainer
#17 lore raw

PATCH v5 11/12 cifs: Remove some code that's no longer used, part 2

In reply to David Howells

Remove some code that was #if'd out with the netfslib conversion. This is split into parts for file.c as the diff generator otherwise produces a hard to read diff for part of it where a big chunk is cut out.

---
 fs/smb/client/file.c | 694 +------------------------------------------
 1 file changed, 1 insertion(+), 693 deletions(-)
fs/smb/client/file.c +1 -693
index e31fd607a441..c50d4071c877 100644
@@ -2563,699 +2563,6 @@ cifs_get_readable_path(struct cifs_tcon *tcon, const char *name,
 	return -ENOENT;
 }
 
-#if 0 // TODO remove 2773
-void
-cifs_writedata_release(struct cifs_io_subrequest *wdata)
-{
-	if (wdata->uncached)
-		kref_put(&wdata->ctx->refcount, cifs_aio_ctx_release);
-#ifdef CONFIG_CIFS_SMB_DIRECT
-	if (wdata->mr) {
-		smbd_deregister_mr(wdata->mr);
-		wdata->mr = NULL;
-	}
-#endif
-
-	if (wdata->cfile)
-		cifsFileInfo_put(wdata->cfile);
-
-	kfree(wdata);
-}
-
-/*
- * Write failed with a retryable error. Resend the write request. It's also
- * possible that the page was redirtied so re-clean the page.
- */
-static void
-cifs_writev_requeue(struct cifs_io_subrequest *wdata)
-{
-	int rc = 0;
-	struct inode *inode = d_inode(wdata->cfile->dentry);
-	struct TCP_Server_Info *server;
-	unsigned int rest_len = wdata->subreq.len;
-	loff_t fpos = wdata->subreq.start;
-
-	server = tlink_tcon(wdata->cfile->tlink)->ses->server;
-	do {
-		struct cifs_io_subrequest *wdata2;
-		unsigned int wsize, cur_len;
-
-		wsize = server->ops->wp_retry_size(inode);
-		if (wsize < rest_len) {
-			if (wsize < PAGE_SIZE) {
-				rc = -EOPNOTSUPP;
-				break;
-			}
-			cur_len = min(round_down(wsize, PAGE_SIZE), rest_len);
-		} else {
-			cur_len = rest_len;
-		}
-
-		wdata2 = cifs_writedata_alloc(cifs_writev_complete);
-		if (!wdata2) {
-			rc = -ENOMEM;
-			break;
-		}
-
-		wdata2->sync_mode = wdata->sync_mode;
-		wdata2->subreq.start	= fpos;
-		wdata2->subreq.len	= cur_len;
-		wdata2->subreq.io_iter = wdata->subreq.io_iter;
-
-		iov_iter_advance(&wdata2->subreq.io_iter, fpos - wdata->subreq.start);
-		iov_iter_truncate(&wdata2->subreq.io_iter, wdata2->subreq.len);
-
-		if (iov_iter_is_xarray(&wdata2->subreq.io_iter))
-			/* Check for pages having been redirtied and clean
-			 * them.  We can do this by walking the xarray.  If
-			 * it's not an xarray, then it's a DIO and we shouldn't
-			 * be mucking around with the page bits.
-			 */
-			cifs_undirty_folios(inode, fpos, cur_len);
-
-		rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY,
-					    &wdata2->cfile);
-		if (!wdata2->cfile) {
-			cifs_dbg(VFS, "No writable handle to retry writepages rc=%d\n",
-				 rc);
-			if (!is_retryable_error(rc))
-				rc = -EBADF;
-		} else {
-			wdata2->pid = wdata2->cfile->pid;
-			rc = server->ops->async_writev(wdata2);
-		}
-
-		cifs_put_writedata(wdata2);
-		if (rc) {
-			if (is_retryable_error(rc))
-				continue;
-			fpos += cur_len;
-			rest_len -= cur_len;
-			break;
-		}
-
-		fpos += cur_len;
-		rest_len -= cur_len;
-	} while (rest_len > 0);
-
-	/* Clean up remaining pages from the original wdata */
-	if (iov_iter_is_xarray(&wdata->subreq.io_iter))
-		cifs_pages_write_failed(inode, fpos, rest_len);
-
-	if (rc != 0 && !is_retryable_error(rc))
-		mapping_set_error(inode->i_mapping, rc);
-	cifs_put_writedata(wdata);
-}
-
-void
-cifs_writev_complete(struct work_struct *work)
-{
-	struct cifs_io_subrequest *wdata = container_of(work,
-						struct cifs_io_subrequest, work);
-	struct inode *inode = d_inode(wdata->cfile->dentry);
-
-	if (wdata->result == 0) {
-		spin_lock(&inode->i_lock);
-		cifs_update_eof(CIFS_I(inode), wdata->subreq.start, wdata->subreq.len);
-		spin_unlock(&inode->i_lock);
-		cifs_stats_bytes_written(tlink_tcon(wdata->cfile->tlink),
-					 wdata->subreq.len);
-	} else if (wdata->sync_mode == WB_SYNC_ALL && wdata->result == -EAGAIN)
-		return cifs_writev_requeue(wdata);
-
-	if (wdata->result == -EAGAIN)
-		cifs_pages_write_redirty(inode, wdata->subreq.start, wdata->subreq.len);
-	else if (wdata->result < 0)
-		cifs_pages_write_failed(inode, wdata->subreq.start, wdata->subreq.len);
-	else
-		cifs_pages_written_back(inode, wdata->subreq.start, wdata->subreq.len);
-
-	if (wdata->result != -EAGAIN)
-		mapping_set_error(inode->i_mapping, wdata->result);
-	cifs_put_writedata(wdata);
-}
-
-struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete)
-{
-	struct cifs_io_subrequest *wdata;
-
-	wdata = kzalloc(sizeof(*wdata), GFP_NOFS);
-	if (wdata != NULL) {
-		refcount_set(&wdata->subreq.ref, 1);
-		INIT_LIST_HEAD(&wdata->list);
-		init_completion(&wdata->done);
-		INIT_WORK(&wdata->work, complete);
-	}
-	return wdata;
-}
-
-static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to)
-{
-	struct address_space *mapping = page->mapping;
-	loff_t offset = (loff_t)page->index << PAGE_SHIFT;
-	char *write_data;
-	int rc = -EFAULT;
-	int bytes_written = 0;
-	struct inode *inode;
-	struct cifsFileInfo *open_file;
-
-	if (!mapping || !mapping->host)
-		return -EFAULT;
-
-	inode = page->mapping->host;
-
-	offset += (loff_t)from;
-	write_data = kmap(page);
-	write_data += from;
-
-	if ((to > PAGE_SIZE) || (from > to)) {
-		kunmap(page);
-		return -EIO;
-	}
-
-	/* racing with truncate? */
-	if (offset > mapping->host->i_size) {
-		kunmap(page);
-		return 0; /* don't care */
-	}
-
-	/* check to make sure that we are not extending the file */
-	if (mapping->host->i_size - offset < (loff_t)to)
-		to = (unsigned)(mapping->host->i_size - offset);
-
-	rc = cifs_get_writable_file(CIFS_I(mapping->host), FIND_WR_ANY,
-				    &open_file);
-	if (!rc) {
-		bytes_written = cifs_write(open_file, open_file->pid,
-					   write_data, to - from, &offset);
-		cifsFileInfo_put(open_file);
-		/* Does mm or vfs already set times? */
-		simple_inode_init_ts(inode);
-		if ((bytes_written > 0) && (offset))
-			rc = 0;
-		else if (bytes_written < 0)
-			rc = bytes_written;
-		else
-			rc = -EFAULT;
-	} else {
-		cifs_dbg(FYI, "No writable handle for write page rc=%d\n", rc);
-		if (!is_retryable_error(rc))
-			rc = -EIO;
-	}
-
-	kunmap(page);
-	return rc;
-}
-
-/*
- * Extend the region to be written back to include subsequent contiguously
- * dirty pages if possible, but don't sleep while doing so.
- */
-static void cifs_extend_writeback(struct address_space *mapping,
-				  long *_count,
-				  loff_t start,
-				  int max_pages,
-				  size_t max_len,
-				  unsigned int *_len)
-{
-	struct folio_batch batch;
-	struct folio *folio;
-	unsigned int psize, nr_pages;
-	size_t len = *_len;
-	pgoff_t index = (start + len) / PAGE_SIZE;
-	bool stop = true;
-	unsigned int i;
-	XA_STATE(xas, &mapping->i_pages, index);
-
-	folio_batch_init(&batch);
-
-	do {
-		/* Firstly, we gather up a batch of contiguous dirty pages
-		 * under the RCU read lock - but we can't clear the dirty flags
-		 * there if any of those pages are mapped.
-		 */
-		rcu_read_lock();
-
-		xas_for_each(&xas, folio, ULONG_MAX) {
-			stop = true;
-			if (xas_retry(&xas, folio))
-				continue;
-			if (xa_is_value(folio))
-				break;
-			if (folio->index != index)
-				break;
-			if (!folio_try_get_rcu(folio)) {
-				xas_reset(&xas);
-				continue;
-			}
-			nr_pages = folio_nr_pages(folio);
-			if (nr_pages > max_pages)
-				break;
-
-			/* Has the page moved or been split? */
-			if (unlikely(folio != xas_reload(&xas))) {
-				folio_put(folio);
-				break;
-			}
-
-			if (!folio_trylock(folio)) {
-				folio_put(folio);
-				break;
-			}
-			if (!folio_test_dirty(folio) || folio_test_writeback(folio)) {
-				folio_unlock(folio);
-				folio_put(folio);
-				break;
-			}
-
-			max_pages -= nr_pages;
-			psize = folio_size(folio);
-			len += psize;
-			stop = false;
-			if (max_pages <= 0 || len >= max_len || *_count <= 0)
-				stop = true;
-
-			index += nr_pages;
-			if (!folio_batch_add(&batch, folio))
-				break;
-			if (stop)
-				break;
-		}
-
-		if (!stop)
-			xas_pause(&xas);
-		rcu_read_unlock();
-
-		/* Now, if we obtained any pages, we can shift them to being
-		 * writable and mark them for caching.
-		 */
-		if (!folio_batch_count(&batch))
-			break;
-
-		for (i = 0; i < folio_batch_count(&batch); i++) {
-			folio = batch.folios[i];
-			/* The folio should be locked, dirty and not undergoing
-			 * writeback from the loop above.
-			 */
-			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;
-}
-
-/*
- * Write back the locked page and any subsequent non-locked dirty pages.
- */
-static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping,
-						 struct writeback_control *wbc,
-						 struct folio *folio,
-						 loff_t start, loff_t end)
-{
-	struct inode *inode = mapping->host;
-	struct TCP_Server_Info *server;
-	struct cifs_io_subrequest *wdata;
-	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
-	struct cifs_credits credits_on_stack;
-	struct cifs_credits *credits = &credits_on_stack;
-	struct cifsFileInfo *cfile = NULL;
-	unsigned int xid, len;
-	loff_t i_size = i_size_read(inode);
-	size_t max_len, wsize;
-	long count = wbc->nr_to_write;
-	int rc;
-
-	/* The folio should be locked, dirty and not undergoing writeback. */
-	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);
-
-	rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY, &cfile);
-	if (rc) {
-		cifs_dbg(VFS, "No writable handle in writepages rc=%d\n", rc);
-		goto err_xid;
-	}
-
-	rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->wsize,
-					   &wsize, credits);
-	if (rc != 0)
-		goto err_close;
-
-	wdata = cifs_writedata_alloc(cifs_writev_complete);
-	if (!wdata) {
-		rc = -ENOMEM;
-		goto err_uncredit;
-	}
-
-	wdata->sync_mode = wbc->sync_mode;
-	wdata->subreq.start = folio_pos(folio);
-	wdata->pid = cfile->pid;
-	wdata->credits = credits_on_stack;
-	wdata->cfile = cfile;
-	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.
-	 */
-	if (start < i_size) {
-		/* Trim the write to the EOF; the extra data is ignored.  Also
-		 * put an upper limit on the size of a single storedata op.
-		 */
-		max_len = wsize;
-		max_len = min_t(unsigned long long, max_len, end - start + 1);
-		max_len = min_t(unsigned long long, max_len, i_size - start);
-
-		if (len < max_len) {
-			int max_pages = INT_MAX;
-
-#ifdef CONFIG_CIFS_SMB_DIRECT
-			if (server->smbd_conn)
-				max_pages = server->smbd_conn->max_frmr_depth;
-#endif
-			max_pages -= folio_nr_pages(folio);
-
-			if (max_pages > 0)
-				cifs_extend_writeback(mapping, &count, start,
-						      max_pages, max_len, &len);
-		}
-		len = min_t(loff_t, len, max_len);
-	}
-
-	wdata->subreq.len = len;
-
-	/* 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);
-
-	if (start < i_size) {
-		iov_iter_xarray(&wdata->subreq.io_iter, ITER_SOURCE, &mapping->i_pages,
-				start, len);
-
-		rc = adjust_credits(wdata->server, &wdata->credits, wdata->subreq.len);
-		if (rc)
-			goto err_wdata;
-
-		if (wdata->cfile->invalidHandle)
-			rc = -EAGAIN;
-		else
-			rc = wdata->server->ops->async_writev(wdata);
-		if (rc >= 0) {
-			cifs_put_writedata(wdata);
-			goto err_close;
-		}
-	} else {
-		/* The dirty region was entirely beyond the EOF. */
-		cifs_pages_written_back(inode, start, len);
-		rc = 0;
-	}
-
-err_wdata:
-	cifs_put_writedata(wdata);
-err_uncredit:
-	add_credits_and_wake_if(server, credits, 0);
-err_close:
-	if (cfile)
-		cifsFileInfo_put(cfile);
-err_xid:
-	free_xid(xid);
-	if (rc == 0) {
-		wbc->nr_to_write = count;
-		rc = len;
-	} else if (is_retryable_error(rc)) {
-		cifs_pages_write_redirty(inode, start, len);
-	} else {
-		cifs_pages_write_failed(inode, start, len);
-		mapping_set_error(mapping, rc);
-	}
-	/* Indication to update ctime and mtime as close is deferred */
-	set_bit(CIFS_INO_MODIFIED_ATTR, &CIFS_I(inode)->flags);
-	return rc;
-}
-
-/*
- * 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)
-{
-	struct folio_batch fbatch;
-	int skips = 0;
-
-	folio_batch_init(&fbatch);
-	do {
-		int nr;
-		pgoff_t index = start / PAGE_SIZE;
-
-		nr = filemap_get_folios_tag(mapping, &index, end / PAGE_SIZE,
-					    PAGECACHE_TAG_DIRTY, &fbatch);
-		if (!nr)
-			break;
-
-		for (int i = 0; i < nr; i++) {
-			ssize_t ret;
-			struct folio *folio = fbatch.folios[i];
-
-redo_folio:
-			start = folio_pos(folio); /* May regress with THPs */
-
-			/* 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;
-			}
-
-			if (folio->mapping != mapping ||
-			    !folio_test_dirty(folio)) {
-				start += folio_size(folio);
-				folio_unlock(folio);
-				continue;
-			}
-
-			if (folio_test_writeback(folio) ||
-			    folio_test_fscache(folio)) {
-				folio_unlock(folio);
-				if (wbc->sync_mode == WB_SYNC_NONE)
-					goto skip_write;
-
-				folio_wait_writeback(folio);
-#ifdef CONFIG_CIFS_FSCACHE
-				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;
-
-skip_write:
-			/*
-			 * Too many skipped writes, or need to reschedule?
-			 * Treat it as a write error without an error code.
-			 */
-			if (skips >= 5 || need_resched()) {
-				ret = 0;
-				goto write_error;
-			}
-
-			/* Otherwise, just skip that folio and go on to the next */
-			skips++;
-			start += folio_size(folio);
-			continue;
-		}
-
-		folio_batch_release(&fbatch);
-		cond_resched();
-	} while (wbc->nr_to_write > 0);
-
-	*_next = start;
-	return 0;
-}
-
-/*
- * Write some of the pending data back to the server
- */
-static int cifs_writepages(struct address_space *mapping,
-			   struct writeback_control *wbc)
-{
-	loff_t start, next;
-	int ret;
-
-	/* We have to be careful as we can end up racing with setattr()
-	 * truncating the pagecache since the caller doesn't take a lock here
-	 * to prevent it.
-	 */
-
-	if (wbc->range_cyclic) {
-		start = mapping->writeback_index * PAGE_SIZE;
-		ret = cifs_writepages_region(mapping, wbc, start, LLONG_MAX, &next);
-		if (ret == 0) {
-			mapping->writeback_index = next / PAGE_SIZE;
-			if (start > 0 && wbc->nr_to_write > 0) {
-				ret = cifs_writepages_region(mapping, wbc, 0,
-							     start, &next);
-				if (ret == 0)
-					mapping->writeback_index =
-						next / PAGE_SIZE;
-			}
-		}
-	} else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) {
-		ret = cifs_writepages_region(mapping, wbc, 0, LLONG_MAX, &next);
-		if (wbc->nr_to_write > 0 && ret == 0)
-			mapping->writeback_index = next / PAGE_SIZE;
-	} else {
-		ret = cifs_writepages_region(mapping, wbc,
-					     wbc->range_start, wbc->range_end, &next);
-	}
-
-	return ret;
-}
-
-static int
-cifs_writepage_locked(struct page *page, struct writeback_control *wbc)
-{
-	int rc;
-	unsigned int xid;
-
-	xid = get_xid();
-/* BB add check for wbc flags */
-	get_page(page);
-	if (!PageUptodate(page))
-		cifs_dbg(FYI, "ppw - page not up to date\n");
-
-	/*
-	 * Set the "writeback" flag, and clear "dirty" in the radix tree.
-	 *
-	 * A writepage() implementation always needs to do either this,
-	 * or re-dirty the page with "redirty_page_for_writepage()" in
-	 * the case of a failure.
-	 *
-	 * Just unlocking the page will cause the radix tree tag-bits
-	 * to fail to update with the state of the page correctly.
-	 */
-	set_page_writeback(page);
-retry_write:
-	rc = cifs_partialpagewrite(page, 0, PAGE_SIZE);
-	if (is_retryable_error(rc)) {
-		if (wbc->sync_mode == WB_SYNC_ALL && rc == -EAGAIN)
-			goto retry_write;
-		redirty_page_for_writepage(wbc, page);
-	} else if (rc != 0) {
-		SetPageError(page);
-		mapping_set_error(page->mapping, rc);
-	} else {
-		SetPageUptodate(page);
-	}
-	end_page_writeback(page);
-	put_page(page);
-	free_xid(xid);
-	return rc;
-}
-
-static int cifs_write_end(struct file *file, struct address_space *mapping,
-			loff_t pos, unsigned len, unsigned copied,
-			struct page *page, void *fsdata)
-{
-	int rc;
-	struct inode *inode = mapping->host;
-	struct cifsFileInfo *cfile = file->private_data;
-	struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb);
-	struct folio *folio = page_folio(page);
-	__u32 pid;
-
-	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
-		pid = cfile->pid;
-	else
-		pid = current->tgid;
-
-	cifs_dbg(FYI, "write_end for page %p from pos %lld with %d bytes\n",
-		 page, pos, copied);
-
-	if (folio_test_checked(folio)) {
-		if (copied == len)
-			folio_mark_uptodate(folio);
-		folio_clear_checked(folio);
-	} else if (!folio_test_uptodate(folio) && copied == PAGE_SIZE)
-		folio_mark_uptodate(folio);
-
-	if (!folio_test_uptodate(folio)) {
-		char *page_data;
-		unsigned offset = pos & (PAGE_SIZE - 1);
-		unsigned int xid;
-
-		xid = get_xid();
-		/* this is probably better than directly calling
-		   partialpage_write since in this function the file handle is
-		   known which we might as well	leverage */
-		/* BB check if anything else missing out of ppw
-		   such as updating last write time */
-		page_data = kmap(page);
-		rc = cifs_write(cfile, pid, page_data + offset, copied, &pos);
-		/* if (rc < 0) should we set writebehind rc? */
-		kunmap(page);
-
-		free_xid(xid);
-	} else {
-		rc = copied;
-		pos += copied;
-		set_page_dirty(page);
-	}
-
-	if (rc > 0) {
-		spin_lock(&inode->i_lock);
-		if (pos > inode->i_size) {
-			i_size_write(inode, pos);
-			inode->i_blocks = (512 - 1 + pos) >> 9;
-		}
-		spin_unlock(&inode->i_lock);
-	}
-
-	unlock_page(page);
-	put_page(page);
-	/* Indication to update ctime and mtime as close is deferred */
-	set_bit(CIFS_INO_MODIFIED_ATTR, &CIFS_I(inode)->flags);
-
-	return rc;
-}
-#endif // End netfs removal 2773
-
 /*
  * Flush data on a strict file.
  */
@@ -4571,6 +3878,7 @@ cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset)
 }
 #endif // end netfslib remove 4633
 
+
 static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf)
 {
 	return netfs_page_mkwrite(vmf, NULL);
author maintainer
#18 lore raw

PATCH v5 12/12 cifs: Remove some code that's no longer used, part 3

In reply to David Howells

Remove some code that was #if'd out with the netfslib conversion. This is split into parts for file.c as the diff generator otherwise produces a hard to read diff for part of it where a big chunk is cut out.

---
 fs/smb/client/file.c | 1004 ------------------------------------------
 1 file changed, 1004 deletions(-)
fs/smb/client/file.c +0 -1004
index c50d4071c877..d3ca2ac3e749 100644
@@ -2689,471 +2689,6 @@ int cifs_flush(struct file *file, fl_owner_t id)
 	return rc;
 }
 
-#if 0 // TODO remove 3594
-static void collect_uncached_write_data(struct cifs_aio_ctx *ctx);
-
-static void
-cifs_uncached_writev_complete(struct work_struct *work)
-{
-	struct cifs_io_subrequest *wdata = container_of(work,
-					struct cifs_io_subrequest, work);
-	struct inode *inode = d_inode(wdata->cfile->dentry);
-	struct cifsInodeInfo *cifsi = CIFS_I(inode);
-
-	spin_lock(&inode->i_lock);
-	cifs_update_eof(cifsi, wdata->subreq.start, wdata->subreq.len);
-	if (cifsi->netfs.remote_i_size > inode->i_size)
-		i_size_write(inode, cifsi->netfs.remote_i_size);
-	spin_unlock(&inode->i_lock);
-
-	complete(&wdata->done);
-	collect_uncached_write_data(wdata->ctx);
-	/* the below call can possibly free the last ref to aio ctx */
-	cifs_put_writedata(wdata);
-}
-
-static int
-cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list,
-	struct cifs_aio_ctx *ctx)
-{
-	size_t wsize;
-	struct cifs_credits credits;
-	int rc;
-	struct TCP_Server_Info *server = wdata->server;
-
-	do {
-		if (wdata->cfile->invalidHandle) {
-			rc = cifs_reopen_file(wdata->cfile, false);
-			if (rc == -EAGAIN)
-				continue;
-			else if (rc)
-				break;
-		}
-
-
-		/*
-		 * Wait for credits to resend this wdata.
-		 * Note: we are attempting to resend the whole wdata not in
-		 * segments
-		 */
-		do {
-			rc = server->ops->wait_mtu_credits(server, wdata->subreq.len,
-						&wsize, &credits);
-			if (rc)
-				goto fail;
-
-			if (wsize < wdata->subreq.len) {
-				add_credits_and_wake_if(server, &credits, 0);
-				msleep(1000);
-			}
-		} while (wsize < wdata->subreq.len);
-		wdata->credits = credits;
-
-		rc = adjust_credits(server, &wdata->credits, wdata->subreq.len);
-
-		if (!rc) {
-			if (wdata->cfile->invalidHandle)
-				rc = -EAGAIN;
-			else {
-				set_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags);
-#ifdef CONFIG_CIFS_SMB_DIRECT
-				if (wdata->mr) {
-					wdata->mr->need_invalidate = true;
-					smbd_deregister_mr(wdata->mr);
-					wdata->mr = NULL;
-				}
-#endif
-				rc = server->ops->async_writev(wdata);
-			}
-		}
-
-		/* If the write was successfully sent, we are done */
-		if (!rc) {
-			list_add_tail(&wdata->list, wdata_list);
-			return 0;
-		}
-
-		/* Roll back credits and retry if needed */
-		add_credits_and_wake_if(server, &wdata->credits, 0);
-	} while (rc == -EAGAIN);
-
-fail:
-	cifs_put_writedata(wdata);
-	return rc;
-}
-
-/*
- * Select span of a bvec iterator we're going to use.  Limit it by both maximum
- * size and maximum number of segments.
- */
-static size_t cifs_limit_bvec_subset(const struct iov_iter *iter, size_t max_size,
-				     size_t max_segs, unsigned int *_nsegs)
-{
-	const struct bio_vec *bvecs = iter->bvec;
-	unsigned int nbv = iter->nr_segs, ix = 0, nsegs = 0;
-	size_t len, span = 0, n = iter->count;
-	size_t skip = iter->iov_offset;
-
-	if (WARN_ON(!iov_iter_is_bvec(iter)) || n == 0)
-		return 0;
-
-	while (n && ix < nbv && skip) {
-		len = bvecs[ix].bv_len;
-		if (skip < len)
-			break;
-		skip -= len;
-		n -= len;
-		ix++;
-	}
-
-	while (n && ix < nbv) {
-		len = min3(n, bvecs[ix].bv_len - skip, max_size);
-		span += len;
-		max_size -= len;
-		nsegs++;
-		ix++;
-		if (max_size == 0 || nsegs >= max_segs)
-			break;
-		skip = 0;
-		n -= len;
-	}
-
-	*_nsegs = nsegs;
-	return span;
-}
-
-static int
-cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from,
-		     struct cifsFileInfo *open_file,
-		     struct cifs_sb_info *cifs_sb, struct list_head *wdata_list,
-		     struct cifs_aio_ctx *ctx)
-{
-	int rc = 0;
-	size_t cur_len, max_len;
-	struct cifs_io_subrequest *wdata;
-	pid_t pid;
-	struct TCP_Server_Info *server;
-	unsigned int xid, max_segs = INT_MAX;
-
-	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
-		pid = open_file->pid;
-	else
-		pid = current->tgid;
-
-	server = cifs_pick_channel(tlink_tcon(open_file->tlink)->ses);
-	xid = get_xid();
-
-#ifdef CONFIG_CIFS_SMB_DIRECT
-	if (server->smbd_conn)
-		max_segs = server->smbd_conn->max_frmr_depth;
-#endif
-
-	do {
-		struct cifs_credits credits_on_stack;
-		struct cifs_credits *credits = &credits_on_stack;
-		unsigned int nsegs = 0;
-		size_t wsize;
-
-		if (signal_pending(current)) {
-			rc = -EINTR;
-			break;
-		}
-
-		if (open_file->invalidHandle) {
-			rc = cifs_reopen_file(open_file, false);
-			if (rc == -EAGAIN)
-				continue;
-			else if (rc)
-				break;
-		}
-
-		rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->wsize,
-						   &wsize, credits);
-		if (rc)
-			break;
-
-		max_len = min_t(const size_t, len, wsize);
-		if (!max_len) {
-			rc = -EAGAIN;
-			add_credits_and_wake_if(server, credits, 0);
-			break;
-		}
-
-		cur_len = cifs_limit_bvec_subset(from, max_len, max_segs, &nsegs);
-		cifs_dbg(FYI, "write_from_iter len=%zx/%zx nsegs=%u/%lu/%u\n",
-			 cur_len, max_len, nsegs, from->nr_segs, max_segs);
-		if (cur_len == 0) {
-			rc = -EIO;
-			add_credits_and_wake_if(server, credits, 0);
-			break;
-		}
-
-		wdata = cifs_writedata_alloc(cifs_uncached_writev_complete);
-		if (!wdata) {
-			rc = -ENOMEM;
-			add_credits_and_wake_if(server, credits, 0);
-			break;
-		}
-
-		wdata->uncached	= true;
-		wdata->sync_mode = WB_SYNC_ALL;
-		wdata->subreq.start	= (__u64)fpos;
-		wdata->cfile	= cifsFileInfo_get(open_file);
-		wdata->server	= server;
-		wdata->pid	= pid;
-		wdata->subreq.len	= cur_len;
-		wdata->credits	= credits_on_stack;
-		wdata->subreq.io_iter	= *from;
-		wdata->ctx	= ctx;
-		kref_get(&ctx->refcount);
-
-		iov_iter_truncate(&wdata->subreq.io_iter, cur_len);
-
-		rc = adjust_credits(server, &wdata->credits, wdata->subreq.len);
-
-		if (!rc) {
-			if (wdata->cfile->invalidHandle)
-				rc = -EAGAIN;
-			else
-				rc = server->ops->async_writev(wdata);
-		}
-
-		if (rc) {
-			add_credits_and_wake_if(server, &wdata->credits, 0);
-			cifs_put_writedata(wdata);
-			if (rc == -EAGAIN)
-				continue;
-			break;
-		}
-
-		list_add_tail(&wdata->list, wdata_list);
-		iov_iter_advance(from, cur_len);
-		fpos += cur_len;
-		len -= cur_len;
-	} while (len > 0);
-
-	free_xid(xid);
-	return rc;
-}
-
-static void collect_uncached_write_data(struct cifs_aio_ctx *ctx)
-{
-	struct cifs_io_subrequest *wdata, *tmp;
-	struct cifs_tcon *tcon;
-	struct cifs_sb_info *cifs_sb;
-	struct dentry *dentry = ctx->cfile->dentry;
-	ssize_t rc;
-
-	tcon = tlink_tcon(ctx->cfile->tlink);
-	cifs_sb = CIFS_SB(dentry->d_sb);
-
-	mutex_lock(&ctx->aio_mutex);
-
-	if (list_empty(&ctx->list)) {
-		mutex_unlock(&ctx->aio_mutex);
-		return;
-	}
-
-	rc = ctx->rc;
-	/*
-	 * Wait for and collect replies for any successful sends in order of
-	 * increasing offset. Once an error is hit, then return without waiting
-	 * for any more replies.
-	 */
-restart_loop:
-	list_for_each_entry_safe(wdata, tmp, &ctx->list, list) {
-		if (!rc) {
-			if (!try_wait_for_completion(&wdata->done)) {
-				mutex_unlock(&ctx->aio_mutex);
-				return;
-			}
-
-			if (wdata->result)
-				rc = wdata->result;
-			else
-				ctx->total_len += wdata->subreq.len;
-
-			/* resend call if it's a retryable error */
-			if (rc == -EAGAIN) {
-				struct list_head tmp_list;
-				struct iov_iter tmp_from = ctx->iter;
-
-				INIT_LIST_HEAD(&tmp_list);
-				list_del_init(&wdata->list);
-
-				if (ctx->direct_io)
-					rc = cifs_resend_wdata(
-						wdata, &tmp_list, ctx);
-				else {
-					iov_iter_advance(&tmp_from,
-						 wdata->subreq.start - ctx->pos);
-
-					rc = cifs_write_from_iter(wdata->subreq.start,
-						wdata->subreq.len, &tmp_from,
-						ctx->cfile, cifs_sb, &tmp_list,
-						ctx);
-
-					cifs_put_writedata(wdata);
-				}
-
-				list_splice(&tmp_list, &ctx->list);
-				goto restart_loop;
-			}
-		}
-		list_del_init(&wdata->list);
-		cifs_put_writedata(wdata);
-	}
-
-	cifs_stats_bytes_written(tcon, ctx->total_len);
-	set_bit(CIFS_INO_INVALID_MAPPING, &CIFS_I(dentry->d_inode)->flags);
-
-	ctx->rc = (rc == 0) ? ctx->total_len : rc;
-
-	mutex_unlock(&ctx->aio_mutex);
-
-	if (ctx->iocb && ctx->iocb->ki_complete)
-		ctx->iocb->ki_complete(ctx->iocb, ctx->rc);
-	else
-		complete(&ctx->done);
-}
-
-static ssize_t __cifs_writev(
-	struct kiocb *iocb, struct iov_iter *from, bool direct)
-{
-	struct file *file = iocb->ki_filp;
-	ssize_t total_written = 0;
-	struct cifsFileInfo *cfile;
-	struct cifs_tcon *tcon;
-	struct cifs_sb_info *cifs_sb;
-	struct cifs_aio_ctx *ctx;
-	int rc;
-
-	rc = generic_write_checks(iocb, from);
-	if (rc <= 0)
-		return rc;
-
-	cifs_sb = CIFS_FILE_SB(file);
-	cfile = file->private_data;
-	tcon = tlink_tcon(cfile->tlink);
-
-	if (!tcon->ses->server->ops->async_writev)
-		return -ENOSYS;
-
-	ctx = cifs_aio_ctx_alloc();
-	if (!ctx)
-		return -ENOMEM;
-
-	ctx->cfile = cifsFileInfo_get(cfile);
-
-	if (!is_sync_kiocb(iocb))
-		ctx->iocb = iocb;
-
-	ctx->pos = iocb->ki_pos;
-	ctx->direct_io = direct;
-	ctx->nr_pinned_pages = 0;
-
-	if (user_backed_iter(from)) {
-		/*
-		 * Extract IOVEC/UBUF-type iterators to a BVEC-type iterator as
-		 * they contain references to the calling process's virtual
-		 * memory layout which won't be available in an async worker
-		 * thread.  This also takes a pin on every folio involved.
-		 */
-		rc = netfs_extract_user_iter(from, iov_iter_count(from),
-					     &ctx->iter, 0);
-		if (rc < 0) {
-			kref_put(&ctx->refcount, cifs_aio_ctx_release);
-			return rc;
-		}
-
-		ctx->nr_pinned_pages = rc;
-		ctx->bv = (void *)ctx->iter.bvec;
-		ctx->bv_need_unpin = iov_iter_extract_will_pin(from);
-	} else if ((iov_iter_is_bvec(from) || iov_iter_is_kvec(from)) &&
-		   !is_sync_kiocb(iocb)) {
-		/*
-		 * If the op is asynchronous, we need to copy the list attached
-		 * to a BVEC/KVEC-type iterator, but we assume that the storage
-		 * will be pinned by the caller; in any case, we may or may not
-		 * be able to pin the pages, so we don't try.
-		 */
-		ctx->bv = (void *)dup_iter(&ctx->iter, from, GFP_KERNEL);
-		if (!ctx->bv) {
-			kref_put(&ctx->refcount, cifs_aio_ctx_release);
-			return -ENOMEM;
-		}
-	} else {
-		/*
-		 * Otherwise, we just pass the iterator down as-is and rely on
-		 * the caller to make sure the pages referred to by the
-		 * iterator don't evaporate.
-		 */
-		ctx->iter = *from;
-	}
-
-	ctx->len = iov_iter_count(&ctx->iter);
-
-	/* grab a lock here due to read response handlers can access ctx */
-	mutex_lock(&ctx->aio_mutex);
-
-	rc = cifs_write_from_iter(iocb->ki_pos, ctx->len, &ctx->iter,
-				  cfile, cifs_sb, &ctx->list, ctx);
-
-	/*
-	 * If at least one write was successfully sent, then discard any rc
-	 * value from the later writes. If the other write succeeds, then
-	 * we'll end up returning whatever was written. If it fails, then
-	 * we'll get a new rc value from that.
-	 */
-	if (!list_empty(&ctx->list))
-		rc = 0;
-
-	mutex_unlock(&ctx->aio_mutex);
-
-	if (rc) {
-		kref_put(&ctx->refcount, cifs_aio_ctx_release);
-		return rc;
-	}
-
-	if (!is_sync_kiocb(iocb)) {
-		kref_put(&ctx->refcount, cifs_aio_ctx_release);
-		return -EIOCBQUEUED;
-	}
-
-	rc = wait_for_completion_killable(&ctx->done);
-	if (rc) {
-		mutex_lock(&ctx->aio_mutex);
-		ctx->rc = rc = -EINTR;
-		total_written = ctx->total_len;
-		mutex_unlock(&ctx->aio_mutex);
-	} else {
-		rc = ctx->rc;
-		total_written = ctx->total_len;
-	}
-
-	kref_put(&ctx->refcount, cifs_aio_ctx_release);
-
-	if (unlikely(!total_written))
-		return rc;
-
-	iocb->ki_pos += total_written;
-	return total_written;
-}
-
-ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from)
-{
-	struct file *file = iocb->ki_filp;
-
-	cifs_revalidate_mapping(file->f_inode);
-	return __cifs_writev(iocb, from, true);
-}
-
-ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from)
-{
-	return __cifs_writev(iocb, from, false);
-}
-#endif // TODO remove 3594
-
 static ssize_t
 cifs_writev(struct kiocb *iocb, struct iov_iter *from)
 {
@@ -3242,450 +2777,6 @@ cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from)
 	return written;
 }
 
-#if 0 // TODO remove 4143
-static struct cifs_io_subrequest *cifs_readdata_alloc(work_func_t complete)
-{
-	struct cifs_io_subrequest *rdata;
-
-	rdata = kzalloc(sizeof(*rdata), GFP_KERNEL);
-	if (rdata) {
-		refcount_set(&rdata->subreq.ref, 1);
-		INIT_LIST_HEAD(&rdata->list);
-		init_completion(&rdata->done);
-		INIT_WORK(&rdata->work, complete);
-	}
-
-	return rdata;
-}
-
-void
-cifs_readdata_release(struct cifs_io_subrequest *rdata)
-{
-	if (rdata->ctx)
-		kref_put(&rdata->ctx->refcount, cifs_aio_ctx_release);
-#ifdef CONFIG_CIFS_SMB_DIRECT
-	if (rdata->mr) {
-		smbd_deregister_mr(rdata->mr);
-		rdata->mr = NULL;
-	}
-#endif
-	if (rdata->cfile)
-		cifsFileInfo_put(rdata->cfile);
-
-	kfree(rdata);
-}
-
-static void collect_uncached_read_data(struct cifs_aio_ctx *ctx);
-
-static void
-cifs_uncached_readv_complete(struct work_struct *work)
-{
-	struct cifs_io_subrequest *rdata =
-		container_of(work, struct cifs_io_subrequest, work);
-
-	complete(&rdata->done);
-	collect_uncached_read_data(rdata->ctx);
-	/* the below call can possibly free the last ref to aio ctx */
-	cifs_put_readdata(rdata);
-}
-
-static int cifs_resend_rdata(struct cifs_io_subrequest *rdata,
-			struct list_head *rdata_list,
-			struct cifs_aio_ctx *ctx)
-{
-	size_t rsize;
-	struct cifs_credits credits;
-	int rc;
-	struct TCP_Server_Info *server;
-
-	/* XXX: should we pick a new channel here? */
-	server = rdata->server;
-
-	do {
-		if (rdata->cfile->invalidHandle) {
-			rc = cifs_reopen_file(rdata->cfile, true);
-			if (rc == -EAGAIN)
-				continue;
-			else if (rc)
-				break;
-		}
-
-		/*
-		 * Wait for credits to resend this rdata.
-		 * Note: we are attempting to resend the whole rdata not in
-		 * segments
-		 */
-		do {
-			rc = server->ops->wait_mtu_credits(server, rdata->subreq.len,
-						&rsize, &credits);
-
-			if (rc)
-				goto fail;
-
-			if (rsize < rdata->subreq.len) {
-				add_credits_and_wake_if(server, &credits, 0);
-				msleep(1000);
-			}
-		} while (rsize < rdata->subreq.len);
-		rdata->credits = credits;
-
-		rc = adjust_credits(server, &rdata->credits, rdata->subreq.len);
-		if (!rc) {
-			if (rdata->cfile->invalidHandle)
-				rc = -EAGAIN;
-			else {
-#ifdef CONFIG_CIFS_SMB_DIRECT
-				if (rdata->mr) {
-					rdata->mr->need_invalidate = true;
-					smbd_deregister_mr(rdata->mr);
-					rdata->mr = NULL;
-				}
-#endif
-				rc = server->ops->async_readv(rdata);
-			}
-		}
-
-		/* If the read was successfully sent, we are done */
-		if (!rc) {
-			/* Add to aio pending list */
-			list_add_tail(&rdata->list, rdata_list);
-			return 0;
-		}
-
-		/* Roll back credits and retry if needed */
-		add_credits_and_wake_if(server, &rdata->credits, 0);
-	} while (rc == -EAGAIN);
-
-fail:
-	cifs_put_readdata(rdata);
-	return rc;
-}
-
-static int
-cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file,
-		     struct cifs_sb_info *cifs_sb, struct list_head *rdata_list,
-		     struct cifs_aio_ctx *ctx)
-{
-	struct cifs_io_subrequest *rdata;
-	unsigned int nsegs, max_segs = INT_MAX;
-	struct cifs_credits credits_on_stack;
-	struct cifs_credits *credits = &credits_on_stack;
-	size_t cur_len, max_len, rsize;
-	int rc;
-	pid_t pid;
-	struct TCP_Server_Info *server;
-
-	server = cifs_pick_channel(tlink_tcon(open_file->tlink)->ses);
-
-#ifdef CONFIG_CIFS_SMB_DIRECT
-	if (server->smbd_conn)
-		max_segs = server->smbd_conn->max_frmr_depth;
-#endif
-
-	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
-		pid = open_file->pid;
-	else
-		pid = current->tgid;
-
-	do {
-		if (open_file->invalidHandle) {
-			rc = cifs_reopen_file(open_file, true);
-			if (rc == -EAGAIN)
-				continue;
-			else if (rc)
-				break;
-		}
-
-		if (cifs_sb->ctx->rsize == 0)
-			cifs_sb->ctx->rsize =
-				server->ops->negotiate_rsize(tlink_tcon(open_file->tlink),
-							     cifs_sb->ctx);
-
-		rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->rsize,
-						   &rsize, credits);
-		if (rc)
-			break;
-
-		max_len = min_t(size_t, len, rsize);
-
-		cur_len = cifs_limit_bvec_subset(&ctx->iter, max_len,
-						 max_segs, &nsegs);
-		cifs_dbg(FYI, "read-to-iter len=%zx/%zx nsegs=%u/%lu/%u\n",
-			 cur_len, max_len, nsegs, ctx->iter.nr_segs, max_segs);
-		if (cur_len == 0) {
-			rc = -EIO;
-			add_credits_and_wake_if(server, credits, 0);
-			break;
-		}
-
-		rdata = cifs_readdata_alloc(cifs_uncached_readv_complete);
-		if (!rdata) {
-			add_credits_and_wake_if(server, credits, 0);
-			rc = -ENOMEM;
-			break;
-		}
-
-		rdata->server	= server;
-		rdata->cfile	= cifsFileInfo_get(open_file);
-		rdata->subreq.start	= fpos;
-		rdata->subreq.len	= cur_len;
-		rdata->pid	= pid;
-		rdata->credits	= credits_on_stack;
-		rdata->ctx	= ctx;
-		kref_get(&ctx->refcount);
-
-		rdata->subreq.io_iter = ctx->iter;
-		iov_iter_truncate(&rdata->subreq.io_iter, cur_len);
-
-		rc = adjust_credits(server, &rdata->credits, rdata->subreq.len);
-
-		if (!rc) {
-			if (rdata->cfile->invalidHandle)
-				rc = -EAGAIN;
-			else
-				rc = server->ops->async_readv(rdata);
-		}
-
-		if (rc) {
-			add_credits_and_wake_if(server, &rdata->credits, 0);
-			cifs_put_readdata(rdata);
-			if (rc == -EAGAIN)
-				continue;
-			break;
-		}
-
-		list_add_tail(&rdata->list, rdata_list);
-		iov_iter_advance(&ctx->iter, cur_len);
-		fpos += cur_len;
-		len -= cur_len;
-	} while (len > 0);
-
-	return rc;
-}
-
-static void
-collect_uncached_read_data(struct cifs_aio_ctx *ctx)
-{
-	struct cifs_io_subrequest *rdata, *tmp;
-	struct cifs_sb_info *cifs_sb;
-	int rc;
-
-	cifs_sb = CIFS_SB(ctx->cfile->dentry->d_sb);
-
-	mutex_lock(&ctx->aio_mutex);
-
-	if (list_empty(&ctx->list)) {
-		mutex_unlock(&ctx->aio_mutex);
-		return;
-	}
-
-	rc = ctx->rc;
-	/* the loop below should proceed in the order of increasing offsets */
-again:
-	list_for_each_entry_safe(rdata, tmp, &ctx->list, list) {
-		if (!rc) {
-			if (!try_wait_for_completion(&rdata->done)) {
-				mutex_unlock(&ctx->aio_mutex);
-				return;
-			}
-
-			if (rdata->result == -EAGAIN) {
-				/* resend call if it's a retryable error */
-				struct list_head tmp_list;
-				unsigned int got_bytes = rdata->got_bytes;
-
-				list_del_init(&rdata->list);
-				INIT_LIST_HEAD(&tmp_list);
-
-				if (ctx->direct_io) {
-					/*
-					 * Re-use rdata as this is a
-					 * direct I/O
-					 */
-					rc = cifs_resend_rdata(
-						rdata,
-						&tmp_list, ctx);
-				} else {
-					rc = cifs_send_async_read(
-						rdata->subreq.start + got_bytes,
-						rdata->subreq.len - got_bytes,
-						rdata->cfile, cifs_sb,
-						&tmp_list, ctx);
-
-					cifs_put_readdata(rdata);
-				}
-
-				list_splice(&tmp_list, &ctx->list);
-
-				goto again;
-			} else if (rdata->result)
-				rc = rdata->result;
-
-			/* if there was a short read -- discard anything left */
-			if (rdata->got_bytes && rdata->got_bytes < rdata->subreq.len)
-				rc = -ENODATA;
-
-			ctx->total_len += rdata->got_bytes;
-		}
-		list_del_init(&rdata->list);
-		cifs_put_readdata(rdata);
-	}
-
-	/* mask nodata case */
-	if (rc == -ENODATA)
-		rc = 0;
-
-	ctx->rc = (rc == 0) ? (ssize_t)ctx->total_len : rc;
-
-	mutex_unlock(&ctx->aio_mutex);
-
-	if (ctx->iocb && ctx->iocb->ki_complete)
-		ctx->iocb->ki_complete(ctx->iocb, ctx->rc);
-	else
-		complete(&ctx->done);
-}
-
-static ssize_t __cifs_readv(
-	struct kiocb *iocb, struct iov_iter *to, bool direct)
-{
-	size_t len;
-	struct file *file = iocb->ki_filp;
-	struct cifs_sb_info *cifs_sb;
-	struct cifsFileInfo *cfile;
-	struct cifs_tcon *tcon;
-	ssize_t rc, total_read = 0;
-	loff_t offset = iocb->ki_pos;
-	struct cifs_aio_ctx *ctx;
-
-	len = iov_iter_count(to);
-	if (!len)
-		return 0;
-
-	cifs_sb = CIFS_FILE_SB(file);
-	cfile = file->private_data;
-	tcon = tlink_tcon(cfile->tlink);
-
-	if (!tcon->ses->server->ops->async_readv)
-		return -ENOSYS;
-
-	if ((file->f_flags & O_ACCMODE) == O_WRONLY)
-		cifs_dbg(FYI, "attempting read on write only file instance\n");
-
-	ctx = cifs_aio_ctx_alloc();
-	if (!ctx)
-		return -ENOMEM;
-
-	ctx->pos	= offset;
-	ctx->direct_io	= direct;
-	ctx->len	= len;
-	ctx->cfile	= cifsFileInfo_get(cfile);
-	ctx->nr_pinned_pages = 0;
-
-	if (!is_sync_kiocb(iocb))
-		ctx->iocb = iocb;
-
-	if (user_backed_iter(to)) {
-		/*
-		 * Extract IOVEC/UBUF-type iterators to a BVEC-type iterator as
-		 * they contain references to the calling process's virtual
-		 * memory layout which won't be available in an async worker
-		 * thread.  This also takes a pin on every folio involved.
-		 */
-		rc = netfs_extract_user_iter(to, iov_iter_count(to),
-					     &ctx->iter, 0);
-		if (rc < 0) {
-			kref_put(&ctx->refcount, cifs_aio_ctx_release);
-			return rc;
-		}
-
-		ctx->nr_pinned_pages = rc;
-		ctx->bv = (void *)ctx->iter.bvec;
-		ctx->bv_need_unpin = iov_iter_extract_will_pin(to);
-		ctx->should_dirty = true;
-	} else if ((iov_iter_is_bvec(to) || iov_iter_is_kvec(to)) &&
-		   !is_sync_kiocb(iocb)) {
-		/*
-		 * If the op is asynchronous, we need to copy the list attached
-		 * to a BVEC/KVEC-type iterator, but we assume that the storage
-		 * will be retained by the caller; in any case, we may or may
-		 * not be able to pin the pages, so we don't try.
-		 */
-		ctx->bv = (void *)dup_iter(&ctx->iter, to, GFP_KERNEL);
-		if (!ctx->bv) {
-			kref_put(&ctx->refcount, cifs_aio_ctx_release);
-			return -ENOMEM;
-		}
-	} else {
-		/*
-		 * Otherwise, we just pass the iterator down as-is and rely on
-		 * the caller to make sure the pages referred to by the
-		 * iterator don't evaporate.
-		 */
-		ctx->iter = *to;
-	}
-
-	if (direct) {
-		rc = filemap_write_and_wait_range(file->f_inode->i_mapping,
-						  offset, offset + len - 1);
-		if (rc) {
-			kref_put(&ctx->refcount, cifs_aio_ctx_release);
-			return -EAGAIN;
-		}
-	}
-
-	/* grab a lock here due to read response handlers can access ctx */
-	mutex_lock(&ctx->aio_mutex);
-
-	rc = cifs_send_async_read(offset, len, cfile, cifs_sb, &ctx->list, ctx);
-
-	/* if at least one read request send succeeded, then reset rc */
-	if (!list_empty(&ctx->list))
-		rc = 0;
-
-	mutex_unlock(&ctx->aio_mutex);
-
-	if (rc) {
-		kref_put(&ctx->refcount, cifs_aio_ctx_release);
-		return rc;
-	}
-
-	if (!is_sync_kiocb(iocb)) {
-		kref_put(&ctx->refcount, cifs_aio_ctx_release);
-		return -EIOCBQUEUED;
-	}
-
-	rc = wait_for_completion_killable(&ctx->done);
-	if (rc) {
-		mutex_lock(&ctx->aio_mutex);
-		ctx->rc = rc = -EINTR;
-		total_read = ctx->total_len;
-		mutex_unlock(&ctx->aio_mutex);
-	} else {
-		rc = ctx->rc;
-		total_read = ctx->total_len;
-	}
-
-	kref_put(&ctx->refcount, cifs_aio_ctx_release);
-
-	if (total_read) {
-		iocb->ki_pos += total_read;
-		return total_read;
-	}
-	return rc;
-}
-
-ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to)
-{
-	return __cifs_readv(iocb, to, true);
-}
-
-ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to)
-{
-	return __cifs_readv(iocb, to, false);
-
-}
-#endif // end netfslib removal 4143
-
 ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter)
 {
 	ssize_t rc;
@@ -3784,101 +2875,6 @@ cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to)
 	return rc;
 }
 
-#if 0 // TODO remove 4633
-static ssize_t
-cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset)
-{
-	int rc = -EACCES;
-	unsigned int bytes_read = 0;
-	unsigned int total_read;
-	unsigned int current_read_size;
-	unsigned int rsize;
-	struct cifs_sb_info *cifs_sb;
-	struct cifs_tcon *tcon;
-	struct TCP_Server_Info *server;
-	unsigned int xid;
-	char *cur_offset;
-	struct cifsFileInfo *open_file;
-	struct cifs_io_parms io_parms = {0};
-	int buf_type = CIFS_NO_BUFFER;
-	__u32 pid;
-
-	xid = get_xid();
-	cifs_sb = CIFS_FILE_SB(file);
-
-	/* FIXME: set up handlers for larger reads and/or convert to async */
-	rsize = min_t(unsigned int, cifs_sb->ctx->rsize, CIFSMaxBufSize);
-
-	if (file->private_data == NULL) {
-		rc = -EBADF;
-		free_xid(xid);
-		return rc;
-	}
-	open_file = file->private_data;
-	tcon = tlink_tcon(open_file->tlink);
-	server = cifs_pick_channel(tcon->ses);
-
-	if (!server->ops->sync_read) {
-		free_xid(xid);
-		return -ENOSYS;
-	}
-
-	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
-		pid = open_file->pid;
-	else
-		pid = current->tgid;
-
-	if ((file->f_flags & O_ACCMODE) == O_WRONLY)
-		cifs_dbg(FYI, "attempting read on write only file instance\n");
-
-	for (total_read = 0, cur_offset = read_data; read_size > total_read;
-	     total_read += bytes_read, cur_offset += bytes_read) {
-		do {
-			current_read_size = min_t(uint, read_size - total_read,
-						  rsize);
-			/*
-			 * For windows me and 9x we do not want to request more
-			 * than it negotiated since it will refuse the read
-			 * then.
-			 */
-			if (!(tcon->ses->capabilities &
-				tcon->ses->server->vals->cap_large_files)) {
-				current_read_size = min_t(uint,
-					current_read_size, CIFSMaxBufSize);
-			}
-			if (open_file->invalidHandle) {
-				rc = cifs_reopen_file(open_file, true);
-				if (rc != 0)
-					break;
-			}
-			io_parms.pid = pid;
-			io_parms.tcon = tcon;
-			io_parms.offset = *offset;
-			io_parms.length = current_read_size;
-			io_parms.server = server;
-			rc = server->ops->sync_read(xid, &open_file->fid, &io_parms,
-						    &bytes_read, &cur_offset,
-						    &buf_type);
-		} while (rc == -EAGAIN);
-
-		if (rc || (bytes_read == 0)) {
-			if (total_read) {
-				break;
-			} else {
-				free_xid(xid);
-				return rc;
-			}
-		} else {
-			cifs_stats_bytes_read(tcon, total_read);
-			*offset += bytes_read;
-		}
-	}
-	free_xid(xid);
-	return total_read;
-}
-#endif // end netfslib remove 4633
-
-
 static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf)
 {
 	return netfs_page_mkwrite(vmf, NULL);
#19 lore raw
In reply to David Howells

This looks promising but I do hit an oops fairly early in xfstest runs with the 12 patches in your series (on 6.8-rc3)

[  228.136056] run fstests generic/306 at 2024-02-07 21:32:16
[  228.573734] ------------[ cut here ]------------
[  228.573737] kernel BUG at lib/iov_iter.c:582!
[  228.573744] invalid opcode: 0000 [#1] PREEMPT SMP PTI
[  228.573748] CPU: 2 PID: 4033 Comm: cifsd Tainted: G            E
  6.8.0-rc3+ #2
[  228.573751] Hardware name: Microsoft Corporation Virtual
Machine/Virtual Machine, BIOS Hyper-V UEFI Release v4.1 11/28/2023
[  228.573752] RIP: 0010:iov_iter_revert+0x114/0x120
[  228.573758] Code: 48 89 78 08 31 c0 31 d2 31 c9 31 f6 31 ff 45 31
c0 c3 cc cc cc cc 0f 0b 31 c0 31 d2 31 c9 31 f6 31 ff 45 31 c0 c3 cc
cc cc cc <0f> 0b 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 90
90 90
[  228.573760] RSP: 0018:ffffb8d005c27d68 EFLAGS: 00010246
[  228.573763] RAX: ffffb8d005c27d98 RBX: ffff99934d039b00 RCX: 0000000000000000
[  228.573764] RDX: 0000000000000000 RSI: 0000000000000200 RDI: 0000000000000004
[  228.573766] RBP: ffffb8d005c27e00 R08: 0000000000000000 R09: 0000000000000000
[  228.573767] R10: 0000000000000000 R11: 0000000000000000 R12: ffff999342736900
[  228.573769] R13: ffff9993a7fd4800 R14: ffff9993a7fd2000 R15: 0000000000000001
[  228.573770] FS:  0000000000000000(0000) GS:ffff9994f2b00000(0000)
knlGS:0000000000000000
[  228.573772] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  228.573774] CR2: 00007fe530dffe38 CR3: 0000000103fc8001 CR4: 00000000003706f0
[  228.573777] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  228.573778] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  228.573779] Call Trace:
[  228.573781]  <TASK>
[  228.573783]  ? show_regs+0x6d/0x80
[  228.573787]  ? die+0x37/0xa0
[  228.573789]  ? do_trap+0xd4/0xf0
[  228.573793]  ? do_error_trap+0x71/0xb0
[  228.573795]  ? iov_iter_revert+0x114/0x120
[  228.573798]  ? exc_invalid_op+0x52/0x80
[  228.573801]  ? iov_iter_revert+0x114/0x120
[  228.573803]  ? asm_exc_invalid_op+0x1b/0x20
[  228.573808]  ? iov_iter_revert+0x114/0x120
[  228.573813]  ? smb2_readv_callback+0x50f/0x5b0 [cifs]
[  228.573874]  cifs_demultiplex_thread+0x46e/0xe40 [cifs]
[  228.573920]  ? __pfx_cifs_demultiplex_thread+0x10/0x10 [cifs]
[  228.573962]  kthread+0xef/0x120
[  228.573966]  ? __pfx_kthread+0x10/0x10
[  228.573969]  ret_from_fork+0x44/0x70
[  228.573972]  ? __pfx_kthread+0x10/0x10
[  228.573975]  ret_from_fork_asm+0x1b/0x30
[  228.573979]  </TASK>
[  228.573980] Modules linked in: cmac(E) nls_utf8(E) cifs(E)

cifs_arc4(E) nls_ucs2_utils(E) cifs_md4(E) netfs(E) snd_seq_dummy(E) snd_hrtimer(E) snd_seq_midi(E) snd_seq_midi_event(E) snd_rawmidi(E) snd_seq(E) snd_seq_device(E) snd_timer(E) snd(E) soundcore(E) sunrpc(E) binfmt_misc(E) intel_rapl_msr(E) intel_rapl_common(E) intel_uncore_frequency_common(E) intel_pmc_core(E) intel_vsec(E) pmt_telemetry(E) pmt_class(E) crct10dif_pclmul(E) polyval_clmulni(E) polyval_generic(E) ghash_clmulni_intel(E) sha512_ssse3(E) sha256_ssse3(E) sha1_ssse3(E) aesni_intel(E) crypto_simd(E) cryptd(E) rapl(E) hyperv_drm(E) drm_shmem_helper(E) drm_kms_helper(E) hv_balloon(E) hyperv_fb(E) vmgenid(E) joydev(E) mac_hid(E) serio_raw(E) msr(E) parport_pc(E) ppdev(E) lp(E) drm(E) parport(E) efi_pstore(E) dmi_sysfs(E) ip_tables(E) x_tables(E) autofs4(E) btrfs(E) blake2b_generic(E) xor(E) raid6_pq(E) libcrc32c(E) hid_generic(E) hid_hyperv(E) hv_storvsc(E) hv_netvsc(E) hid(E) hyperv_keyboard(E) scsi_transport_fc(E) hv_utils(E) crc32_pclmul(E)

[  228.574031]  hv_vmbus(E)
[  228.574035] ---[ end trace 0000000000000000 ]---
[  228.636462] RIP: 0010:iov_iter_revert+0x114/0x120
[  228.636471] Code: 48 89 78 08 31 c0 31 d2 31 c9 31 f6 31 ff 45 31
c0 c3 cc cc cc cc 0f 0b 31 c0 31 d2 31 c9 31 f6 31 ff 45 31 c0 c3 cc
cc cc cc <0f> 0b 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 90
90 90
[  228.636474] RSP: 0018:ffffb8d005c27d68 EFLAGS: 00010246
[  228.636477] RAX: ffffb8d005c27d98 RBX: ffff99934d039b00 RCX: 0000000000000000
[  228.636479] RDX: 0000000000000000 RSI: 0000000000000200 RDI: 0000000000000004
[  228.636481] RBP: ffffb8d005c27e00 R08: 0000000000000000 R09: 0000000000000000
[  228.636482] R10: 0000000000000000 R11: 0000000000000000 R12: ffff999342736900
[  228.636484] R13: ffff9993a7fd4800 R14: ffff9993a7fd2000 R15: 0000000000000001
[  228.636485] FS:  0000000000000000(0000) GS:ffff9994f2b00000(0000)
knlGS:0000000000000000
[  228.636487] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  228.636489] CR2: 00007fe530dffe38 CR3: 0000000103fc8001 CR4: 00000000003706f0
[  228.636492] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  228.636494] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Quoted 121 lines · David Howells

Hi Steve,

Here are patches to convert cifs to use the netfslib library. With this I can run a certain amount of xfstests on CIFS, though not all the tests work correctly because of fallocate issues.

The patches remove around 2000 lines from CIFS

To do:

(*) Implement write-retry. Currently, netfslib errors out on a failed DIO
write and relies on the VM to drive writepages again on a failed buffered write. This needs some retry logic adding into fs/netfs/output.c.

I'm not sure what the best way to handle this is. One way is to resend each failing subreq as it fails, offloading this to a kernel thread that re-splits the subreq, calling out to a rreq->op to do the splitting, thereby allowing cifs to renegotiate credits. If a subreq is split, the two parts need to be adjacent in the rreq->subrequests list.

An alternative way might be to try and combine failing tests and then split them.

Yet a third way might be to try each failing subreq a smaller bit at a time and keep track of what has been sent in wdata->subreq.transferred.

Whichever way is chosen, NETFS_SREQ_RETRYING should be set in wdata->subreq.flags instead of setting wdata->replay.

Notes:

(1) CIFS is made to use unbuffered I/O for unbuffered caching modes and write-through caching for cache=strict.

(2) Various cifs fallocate() function implementations have issues that aren't easily fixed without enhanced protocol support.

(3) It should be possible to turn on multipage folio support in CIFS now.

(4) The then-unused CIFS code is removed in three patches, not one, to avoid the git patch generator from producing confusing patches in which it thinks code is being moved around rather than just being removed.

The patches can be found here also:

https://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git/log/?h=cifs-netfs

Changes
=======
ver #5)
- Rebased to -rc3 plus SteveF's for-next branch as netfslib is now upstream, as are a couple of patches from this series.
- Replace the ->replay bool Shyam added with a flag on the netfs subrequest. This is tested by the code, but not currently set (see above).

ver #4)
- Slimmed down the branch:
- Split the cifs-related patches off to a separate branch (cifs-netfs)
- Deferred the content-encryption to the in-progress ceph changes.
- Deferred the use-PG_writeback rather than PG_fscache patch
- Rebased on a later linux-next with afs-rotation patches.

ver #3)
- Moved the fscache module into netfslib to avoid export cycles.
- Fixed a bunch of bugs.
- Got CIFS to pass as much of xfstests as possible.
- Added a patch to make 9P use all the helpers.
- Added a patch to stop using PG_fscache, but rather dirty pages on reading and have writepages write to the cache.

ver #2)
- Folded the addition of NETFS_RREQ_NONBLOCK/BLOCKED into first patch that uses them.
- Folded addition of rsize member into first user.
- Don't set rsize in ceph (yet) and set it in kafs to 256KiB. cifs sets it dynamically.
- Moved direct_bv next to direct_bv_count in struct netfs_io_request and labelled it with a __counted_by().
- Passed flags into netfs_xa_store_and_mark() rather than two bools.
- Removed netfs_set_up_buffer() as it wasn't used.

David

David Howells (12):
cifs: Replace cifs_readdata with a wrapper around netfs_io_subrequest
cifs: Set zero_point in the copy_file_range() and remap_file_range()
cifs: Replace cifs_writedata with a wrapper around netfs_io_subrequest
cifs: Use more fields from netfs_io_subrequest
cifs: Make wait_mtu_credits take size_t args
cifs: Implement netfslib hooks
cifs: Replace the writedata replay bool with a netfs sreq flag
cifs: Move cifs_loose_read_iter() and cifs_file_write_iter() to file.c
cifs: Cut over to using netfslib
cifs: Remove some code that's no longer used, part 1
cifs: Remove some code that's no longer used, part 2
cifs: Remove some code that's no longer used, part 3

 fs/netfs/buffered_write.c    |    3 +
 fs/netfs/io.c                |    7 +-
 fs/smb/client/Kconfig        |    1 +
 fs/smb/client/cifsfs.c       |   69 +-
 fs/smb/client/cifsfs.h       |   10 +-
 fs/smb/client/cifsglob.h     |   59 +-
 fs/smb/client/cifsproto.h    |   14 +-
 fs/smb/client/cifssmb.c      |  111 +-
 fs/smb/client/file.c         | 2911 ++++++----------------------------
 fs/smb/client/fscache.c      |  109 --
 fs/smb/client/fscache.h      |   54 -
 fs/smb/client/inode.c        |   19 +-
Quoted 10 lines
 fs/smb/client/smb2ops.c      |   10 +-
 fs/smb/client/smb2pdu.c      |  169 +-
 fs/smb/client/smb2proto.h    |    5 +-
 fs/smb/client/trace.h        |  144 +-
 fs/smb/client/transport.c    |   17 +-
 include/linux/netfs.h        |    2 +
 include/trace/events/netfs.h |    1 +
 19 files changed, 852 insertions(+), 2863 deletions(-)

--
Thanks,

Steve

reviewer
#10 lore raw

Re: [PATCH v5 09/12] cifs: Cut over to using netfslib

In reply to David Howells

On Mon, Feb 05, 2024 at 10:57:21PM +0000, David Howells wrote:

...

Quoted 9 lines
fs/smb/client/smb2pdu.c +0 -1
index 84e3675eb41e..b58fdee40755 100644
@@ -4386,10 +4386,12 @@ smb2_new_read_req(void **buf, unsigned int *total_len,
 	req->Length = cpu_to_le32(io_parms->length);
 	req->Offset = cpu_to_le64(io_parms->offset);
 
-	trace_smb3_read_enter(0 /* xid */,

fs/smb/client/smb2pdu.c smb2_new_read_req(void **buf, unsigned int *total_len, · line 4395

Quoted 10 lines
fs/smb/client/smb2pdu.c +6 -3
-			io_parms->persistent_fid,
-			io_parms->tcon->tid, io_parms->tcon->ses->Suid,
-			io_parms->offset, io_parms->length);
+	trace_smb3_read_enter(rdata ? rdata->rreq->debug_id : 0,
+			      rdata ? rdata->subreq.debug_index : 0,
+			      rdata ? rdata->xid : 0,
+			      io_parms->persistent_fid,
+			      io_parms->tcon->tid, io_parms->tcon->ses->Suid,
+			      io_parms->offset, io_parms->length);
 #ifdef CONFIG_CIFS_SMB_DIRECT

Hi David,

above some care is taken to handle the case where rdata might be NULL.

However, the code below this hunk, other than being guarded by smb3_use_rdma_offload(io_parms), uses rdata unconditionally.

Perhaps the guard makes this ok. But Smatch flags this inconsistency. And I thought I should bring it to your attention.

For reference the code I am referring to looks like this:

#ifdef CONFIG_CIFS_SMB_DIRECT
	/*
	 * If we want to do a RDMA write, fill in and append
	 * smbd_buffer_descriptor_v1 to the end of read request
	 */
	if (smb3_use_rdma_offload(io_parms)) {
		struct smbd_buffer_descriptor_v1 *v1;
		bool need_invalidate = server->dialect == SMB30_PROT_ID;

		rdata->mr = smbd_register_mr(server->smbd_conn, &rdata->subreq.io_iter,
					     true, need_invalidate);
		if (!rdata->mr)
			return -EAGAIN;

...

reviewer
#12 lore raw

Re: [PATCH v5 09/12] cifs: Cut over to using netfslib

In reply to David Howells

On Mon, Feb 05, 2024 at 10:57:21PM +0000, David Howells wrote:

...

fs/smb/client/cifsfs.h extern const struct inode_operations cifs_namespace_inode_operations; · line 91

Quoted 12 lines
fs/smb/client/cifsfs.h +1 -0
index e8e0f863e935..5cd547b7b5ea 100644
@@ -85,6 +85,7 @@ extern const struct inode_operations cifs_namespace_inode_operations;
 
 
 /* Functions related to files and directories */
+extern const struct netfs_request_ops cifs_req_ops;
 extern const struct file_operations cifs_file_ops;
 extern const struct file_operations cifs_file_direct_ops; /* if directio mnt */
 extern const struct file_operations cifs_file_strict_ops; /* if strictio mnt */

Nit: this hunk would probably be better placed in the
patch at adds cifs_req_ops to fs/smb/client/file.c

...

author maintainer
#11 lore raw

Re: [PATCH v5 09/12] cifs: Cut over to using netfslib

In reply to Simon Horman
Quoted 2 lines · Simon Horman

However, the code below this hunk, other than being guarded by smb3_use_rdma_offload(io_parms), uses rdata unconditionally.

Yeah - it does that even without my patches. SMB2_read() can call the function with rdata == NULL, but I'm unsure as to whether the RDMA branch will ever be used except for actual reads - in which case rdata will not be NULL at the "rdata->mr" point.

David

author maintainer
#13 lore raw

Re: [PATCH v5 09/12] cifs: Cut over to using netfslib

In reply to Simon Horman
Quoted 8 lines · Simon Horman
Quoted 5 lines

/* Functions related to files and directories */
+extern const struct netfs_request_ops cifs_req_ops;
extern const struct file_operations cifs_file_ops;
extern const struct file_operations cifs_file_direct_ops; /* if directio mnt */ extern const struct file_operations cifs_file_strict_ops; /* if strictio mnt */

Nit: this hunk would probably be better placed in the
patch at adds cifs_req_ops to fs/smb/client/file.c

I'm not sure I understand what you mean. Is there a bit missing between "at" and "adds" in that?

David

author maintainer
#20 lore raw
In reply to Steve French
Quoted 11 lines · Steve French
[  228.573783]  ? show_regs+0x6d/0x80
[  228.573787]  ? die+0x37/0xa0
[  228.573789]  ? do_trap+0xd4/0xf0
[  228.573793]  ? do_error_trap+0x71/0xb0
[  228.573795]  ? iov_iter_revert+0x114/0x120
[  228.573798]  ? exc_invalid_op+0x52/0x80
[  228.573801]  ? iov_iter_revert+0x114/0x120
[  228.573803]  ? asm_exc_invalid_op+0x1b/0x20
[  228.573808]  ? iov_iter_revert+0x114/0x120
[  228.573813]  ? smb2_readv_callback+0x50f/0x5b0 [cifs]
[  228.573874]  cifs_demultiplex_thread+0x46e/0xe40 [cifs]

I don't suppose you can tell me what line smb2_readv_callback+0x50f/0x5b0 is?

David

author maintainer
#21 lore raw
Quoted 1 line · David Howells

I don't suppose you can tell me what line smb2_readv_callback+0x50f/0x5b0 is?

It's almost certainly the iov_iter_revert() here:

	switch (mid->mid_state) {
	case MID_RESPONSE_RECEIVED:
		credits.value = le16_to_cpu(shdr->CreditRequest);
		credits.instance = server->reconnect_instance;
		/* result already set, check signature */
		if (server->sign && !mid->decrypted) {
			int rc;

			iov_iter_revert(&rqst.rq_iter, rdata->got_bytes);
			iov_iter_truncate(&rqst.rq_iter, rdata->got_bytes);

The reason that the:

	[  228.573737] kernel BUG at lib/iov_iter.c:582!

happens is that we're trying to wind the iterator back before its start point.

Now, the iterator is reinitialised at the beginning of the function:

	if (rdata->got_bytes) {
		rqst.rq_iter	  = rdata->subreq.io_iter;
		rqst.rq_iter_size = iov_iter_count(&rdata->subreq.io_iter);
	}

so the reversion is probably unnecessary.

Note that this can only happen if we're using signed messages:

		if (server->sign && !mid->decrypted) {

as we wind back the iterator so that we can use it to feed the buffer to the hashing algorithm.

David

author maintainer
#22 lore raw
Quoted 1 line · David Howells

so the reversion is probably unnecessary.

Removing the iov_iter_revert() fixes the problem.

David

author maintainer
#15 lore raw

Re: [PATCH v5 09/12] cifs: Cut over to using netfslib

In reply to Simon Horman
Quoted 2 lines · Simon Horman

Nit: this hunk would probably be better placed in the
patch at adds cifs_req_ops to fs/smb/client/file.c

I've moved that to the patch that adds cifs_req_ops.

David

reviewer
#14 lore raw

Re: [PATCH v5 09/12] cifs: Cut over to using netfslib

In reply to David Howells
Quoted 8 lines · David Howells
Quoted 6 lines · Simon Horman
Quoted 5 lines

/* Functions related to files and directories */
+extern const struct netfs_request_ops cifs_req_ops;
extern const struct file_operations cifs_file_ops;
extern const struct file_operations cifs_file_direct_ops; /* if directio mnt */ extern const struct file_operations cifs_file_strict_ops; /* if strictio mnt */

Quoted 5 lines
Quoted 2 lines

Nit: this hunk would probably be better placed in the
patch at adds cifs_req_ops to fs/smb/client/file.c

I'm not sure I understand what you mean. Is there a bit missing between "at" and "adds" in that?

Sorry, "patch that adds".

What I meant is, the declaration of cifs_req_ops and it's definition seem to appear in different patches of this series. And it might make sense if they were both in the same patch. But given that both are present by the end of the series it is more cosmetic than anything else.