Merge tag 'nfs-for-6.14-1' of git://git.linux-nfs.org/projects/anna/l…

…inux-nfs

Pull NFS client updates from Anna Schumaker:
 "New Features:
   - Enable using direct IO with localio
   - Added localio related tracepoints

  Bugfixes:
   - Sunrpc fixes for working with a very large cl_tasks list
   - Fix a possible buffer overflow in nfs_sysfs_link_rpc_client()
   - Fixes for handling reconnections with localio
   - Fix how the NFS_FSCACHE kconfig option interacts with NETFS_SUPPORT
   - Fix COPY_NOTIFY xdr_buf size calculations
   - pNFS/Flexfiles fix for retrying requesting a layout segment for
     reads
   - Sunrpc fix for retrying on EKEYEXPIRED error when the TGT is
     expired

  Cleanups:
   - Various other nfs & nfsd localio cleanups
   - Prepratory patches for async copy improvements that are under
     development
   - Make OFFLOAD_CANCEL, LAYOUTSTATS, and LAYOUTERR moveable to other
     xprts
   - Add netns inum and srcaddr to debugfs rpc_xprt info"

* tag 'nfs-for-6.14-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (28 commits)
  SUNRPC: do not retry on EKEYEXPIRED when user TGT ticket expired
  sunrpc: add netns inum and srcaddr to debugfs rpc_xprt info
  pnfs/flexfiles: retry getting layout segment for reads
  NFSv4.2: make LAYOUTSTATS and LAYOUTERROR MOVEABLE
  NFSv4.2: mark OFFLOAD_CANCEL MOVEABLE
  NFSv4.2: fix COPY_NOTIFY xdr buf size calculation
  NFS: Rename struct nfs4_offloadcancel_data
  NFS: Fix typo in OFFLOAD_CANCEL comment
  NFS: CB_OFFLOAD can return NFS4ERR_DELAY
  nfs: Make NFS_FSCACHE select NETFS_SUPPORT instead of depending on it
  nfs: fix incorrect error handling in LOCALIO
  nfs: probe for LOCALIO when v3 client reconnects to server
  nfs: probe for LOCALIO when v4 client reconnects to server
  nfs/localio: remove redundant code and simplify LOCALIO enablement
  nfs_common: add nfs_localio trace events
  nfs_common: track all open nfsd_files per LOCALIO nfs_client
  nfs_common: rename nfslocalio nfs_uuid_lock to nfs_uuids_lock
  nfsd: nfsd_file_acquire_local no longer returns GC'd nfsd_file
  nfsd: rename nfsd_serv_ prefixed methods and variables with nfsd_net_
  nfsd: update percpu_ref to manage references on nfsd_net
  ...

Documentation/filesystems/nfs/localio.rst

@@ -218,64 +218,30 @@ NFS Client and Server Interlock
 ===============================
 
 LOCALIO provides the nfs_uuid_t object and associated interfaces to
-allow proper network namespace (net-ns) and NFSD object refcounting:
-
-    We don't want to keep a long-term counted reference on each NFSD's
-    net-ns in the client because that prevents a server container from
-    completely shutting down.
-
-    So we avoid taking a reference at all and rely on the per-cpu
-    reference to the server (detailed below) being sufficient to keep
-    the net-ns active. This involves allowing the NFSD's net-ns exit
-    code to iterate all active clients and clear their ->net pointers
-    (which are needed to find the per-cpu-refcount for the nfsd_serv).
-
-    Details:
-
-     - Embed nfs_uuid_t in nfs_client. nfs_uuid_t provides a list_head
-       that can be used to find the client. It does add the 16-byte
-       uuid_t to nfs_client so it is bigger than needed (given that
-       uuid_t is only used during the initial NFS client and server
-       LOCALIO handshake to determine if they are local to each other).
-       If that is really a problem we can find a fix.
-
-     - When the nfs server confirms that the uuid_t is local, it moves
-       the nfs_uuid_t onto a per-net-ns list in NFSD's nfsd_net.
-
-     - When each server's net-ns is shutting down - in a "pre_exit"
-       handler, all these nfs_uuid_t have their ->net cleared. There is
-       an rcu_synchronize() call between pre_exit() handlers and exit()
-       handlers so any caller that sees nfs_uuid_t ->net as not NULL can
-       safely manage the per-cpu-refcount for nfsd_serv.
-
-     - The client's nfs_uuid_t is passed to nfsd_open_local_fh() so it
-       can safely dereference ->net in a private rcu_read_lock() section
-       to allow safe access to the associated nfsd_net and nfsd_serv.
-
-So LOCALIO required the introduction and use of NFSD's percpu_ref to
-interlock nfsd_destroy_serv() and nfsd_open_local_fh(), to ensure each
-nn->nfsd_serv is not destroyed while in use by nfsd_open_local_fh(), and
+allow proper network namespace (net-ns) and NFSD object refcounting.
+
+LOCALIO required the introduction and use of NFSD's percpu nfsd_net_ref
+to interlock nfsd_shutdown_net() and nfsd_open_local_fh(), to ensure
+each net-ns is not destroyed while in use by nfsd_open_local_fh(), and
 warrants a more detailed explanation:
 
-    nfsd_open_local_fh() uses nfsd_serv_try_get() before opening its
+    nfsd_open_local_fh() uses nfsd_net_try_get() before opening its
     nfsd_file handle and then the caller (NFS client) must drop the
-    reference for the nfsd_file and associated nn->nfsd_serv using
-    nfs_file_put_local() once it has completed its IO.
+    reference for the nfsd_file and associated net-ns using
+    nfsd_file_put_local() once it has completed its IO.
 
     This interlock working relies heavily on nfsd_open_local_fh() being
     afforded the ability to safely deal with the possibility that the
     NFSD's net-ns (and nfsd_net by association) may have been destroyed
-    by nfsd_destroy_serv() via nfsd_shutdown_net() -- which is only
-    possible given the nfs_uuid_t ->net pointer managemenet detailed
-    above.
-
-All told, this elaborate interlock of the NFS client and server has been
-verified to fix an easy to hit crash that would occur if an NFSD
-instance running in a container, with a LOCALIO client mounted, is
-shutdown. Upon restart of the container and associated NFSD the client
-would go on to crash due to NULL pointer dereference that occurred due
-to the LOCALIO client's attempting to nfsd_open_local_fh(), using
-nn->nfsd_serv, without having a proper reference on nn->nfsd_serv.
+    by nfsd_destroy_serv() via nfsd_shutdown_net().
+
+This interlock of the NFS client and server has been verified to fix an
+easy to hit crash that would occur if an NFSD instance running in a
+container, with a LOCALIO client mounted, is shutdown. Upon restart of
+the container and associated NFSD, the client would go on to crash due
+to NULL pointer dereference that occurred due to the LOCALIO client's
+attempting to nfsd_open_local_fh() without having a proper reference on
+NFSD's net-ns.
 
 NFS Client issues IO instead of Server
 ======================================
@@ -306,10 +272,26 @@ is issuing IO to the underlying local filesystem that it is sharing with
 the NFS server. See: fs/nfs/localio.c:nfs_local_doio() and
 fs/nfs/localio.c:nfs_local_commit().
 
+With normal NFS that makes use of RPC to issue IO to the server, if an
+application uses O_DIRECT the NFS client will bypass the pagecache but
+the NFS server will not. The NFS server's use of buffered IO affords
+applications to be less precise with their alignment when issuing IO to
+the NFS client. But if all applications properly align their IO, LOCALIO
+can be configured to use end-to-end O_DIRECT semantics from the NFS
+client to the underlying local filesystem, that it is sharing with
+the NFS server, by setting the 'localio_O_DIRECT_semantics' nfs module
+parameter to Y, e.g.:
+
+    echo Y > /sys/module/nfs/parameters/localio_O_DIRECT_semantics
+
+Once enabled, it will cause LOCALIO to use end-to-end O_DIRECT semantics
+(but again, this may cause IO to fail if applications do not properly
+align their IO).
+
 Security
 ========
 
-Localio is only supported when UNIX-style authentication (AUTH_UNIX, aka
+LOCALIO is only supported when UNIX-style authentication (AUTH_UNIX, aka
 AUTH_SYS) is used.
 
 Care is taken to ensure the same NFS security mechanisms are used
@@ -324,6 +306,24 @@ client is afforded this same level of access (albeit in terms of the NFS
 protocol via SUNRPC). No other namespaces (user, mount, etc) have been
 altered or purposely extended from the server to the client.
 
+Module Parameters
+=================
+
+/sys/module/nfs/parameters/localio_enabled (bool)
+controls if LOCALIO is enabled, defaults to Y. If client and server are
+local but 'localio_enabled' is set to N then LOCALIO will not be used.
+
+/sys/module/nfs/parameters/localio_O_DIRECT_semantics (bool)
+controls if O_DIRECT extends down to the underlying filesystem, defaults
+to N. Application IO must be logical blocksize aligned, otherwise
+O_DIRECT will fail.
+
+/sys/module/nfsv3/parameters/nfs3_localio_probe_throttle (uint)
+controls if NFSv3 read and write IOs will trigger (re)enabling of
+LOCALIO every N (nfs3_localio_probe_throttle) IOs, defaults to 0
+(disabled). Must be power-of-2, admin keeps all the pieces if they
+misconfigure (too low a value or non-power-of-2).
+
 Testing
 =======
 

fs/nfs/Kconfig

@@ -170,7 +170,8 @@ config ROOT_NFS
 
 config NFS_FSCACHE
 	bool "Provide NFS client caching support"
-	depends on NFS_FS=m && NETFS_SUPPORT || NFS_FS=y && NETFS_SUPPORT=y
+	depends on NFS_FS
+	select NETFS_SUPPORT
 	select FSCACHE
 	help
 	  Say Y here if you want NFS data to be cached locally on disc through

fs/nfs/callback_proc.c

@@ -718,7 +718,7 @@ __be32 nfs4_callback_offload(void *data, void *dummy,
 
 	copy = kzalloc(sizeof(struct nfs4_copy_state), GFP_KERNEL);
 	if (!copy)
-		return htonl(NFS4ERR_SERVERFAULT);
+		return cpu_to_be32(NFS4ERR_DELAY);
 
 	spin_lock(&cps->clp->cl_lock);
 	rcu_read_lock();

fs/nfs/client.c

@@ -38,7 +38,7 @@
 #include <linux/sunrpc/bc_xprt.h>
 #include <linux/nsproxy.h>
 #include <linux/pid_namespace.h>
-
+#include <linux/nfslocalio.h>
 
 #include "nfs4_fs.h"
 #include "callback.h"
@@ -186,7 +186,7 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init)
 	seqlock_init(&clp->cl_boot_lock);
 	ktime_get_real_ts64(&clp->cl_nfssvc_boot);
 	nfs_uuid_init(&clp->cl_uuid);
-	spin_lock_init(&clp->cl_localio_lock);
+	INIT_WORK(&clp->cl_local_probe_work, nfs_local_probe_async_work);
 #endif /* CONFIG_NFS_LOCALIO */
 
 	clp->cl_principal = "*";
@@ -244,7 +244,7 @@ static void pnfs_init_server(struct nfs_server *server)
  */
 void nfs_free_client(struct nfs_client *clp)
 {
-	nfs_local_disable(clp);
+	nfs_localio_disable_client(clp);
 
 	/* -EIO all pending I/O */
 	if (!IS_ERR(clp->cl_rpcclient))

fs/nfs/direct.c

@@ -303,6 +303,7 @@ static void nfs_read_sync_pgio_error(struct list_head *head, int error)
 static void nfs_direct_pgio_init(struct nfs_pgio_header *hdr)
 {
 	get_dreq(hdr->dreq);
+	set_bit(NFS_IOHDR_ODIRECT, &hdr->flags);
 }
 
 static const struct nfs_pgio_completion_ops nfs_direct_read_completion_ops = {

fs/nfs/flexfilelayout/flexfilelayout.c

@@ -164,18 +164,17 @@ decode_name(struct xdr_stream *xdr, u32 *id)
 }
 
 static struct nfsd_file *
-ff_local_open_fh(struct nfs_client *clp, const struct cred *cred,
+ff_local_open_fh(struct pnfs_layout_segment *lseg, u32 ds_idx,
+		 struct nfs_client *clp, const struct cred *cred,
 		 struct nfs_fh *fh, fmode_t mode)
 {
-	if (mode & FMODE_WRITE) {
-		/*
-		 * Always request read and write access since this corresponds
-		 * to a rw layout.
-		 */
-		mode |= FMODE_READ;
-	}
+#if IS_ENABLED(CONFIG_NFS_LOCALIO)
+	struct nfs4_ff_layout_mirror *mirror = FF_LAYOUT_COMP(lseg, ds_idx);
 
-	return nfs_local_open_fh(clp, cred, fh, mode);
+	return nfs_local_open_fh(clp, cred, fh, &mirror->nfl, mode);
+#else
+	return NULL;
+#endif
 }
 
 static bool ff_mirror_match_fh(const struct nfs4_ff_layout_mirror *m1,
@@ -247,6 +246,7 @@ static struct nfs4_ff_layout_mirror *ff_layout_alloc_mirror(gfp_t gfp_flags)
 		spin_lock_init(&mirror->lock);
 		refcount_set(&mirror->ref, 1);
 		INIT_LIST_HEAD(&mirror->mirrors);
+		nfs_localio_file_init(&mirror->nfl);
 	}
 	return mirror;
 }
@@ -257,6 +257,7 @@ static void ff_layout_free_mirror(struct nfs4_ff_layout_mirror *mirror)
 
 	ff_layout_remove_mirror(mirror);
 	kfree(mirror->fh_versions);
+	nfs_close_local_fh(&mirror->nfl);
 	cred = rcu_access_pointer(mirror->ro_cred);
 	put_cred(cred);
 	cred = rcu_access_pointer(mirror->rw_cred);
@@ -847,6 +848,9 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 	struct nfs4_pnfs_ds *ds;
 	u32 ds_idx;
 
+	if (NFS_SERVER(pgio->pg_inode)->flags &
+			(NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR))
+		pgio->pg_maxretrans = io_maxretrans;
 retry:
 	pnfs_generic_pg_check_layout(pgio, req);
 	/* Use full layout for now */
@@ -860,6 +864,8 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 		if (!pgio->pg_lseg)
 			goto out_nolseg;
 	}
+	/* Reset wb_nio, since getting layout segment was successful */
+	req->wb_nio = 0;
 
 	ds = ff_layout_get_ds_for_read(pgio, &ds_idx);
 	if (!ds) {
@@ -876,14 +882,24 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 	pgm->pg_bsize = mirror->mirror_ds->ds_versions[0].rsize;
 
 	pgio->pg_mirror_idx = ds_idx;
-
-	if (NFS_SERVER(pgio->pg_inode)->flags &
-			(NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR))
-		pgio->pg_maxretrans = io_maxretrans;
 	return;
 out_nolseg:
-	if (pgio->pg_error < 0)
-		return;
+	if (pgio->pg_error < 0) {
+		if (pgio->pg_error != -EAGAIN)
+			return;
+		/* Retry getting layout segment if lower layer returned -EAGAIN */
+		if (pgio->pg_maxretrans && req->wb_nio++ > pgio->pg_maxretrans) {
+			if (NFS_SERVER(pgio->pg_inode)->flags & NFS_MOUNT_SOFTERR)
+				pgio->pg_error = -ETIMEDOUT;
+			else
+				pgio->pg_error = -EIO;
+			return;
+		}
+		pgio->pg_error = 0;
+		/* Sleep for 1 second before retrying */
+		ssleep(1);
+		goto retry;
+	}
 out_mds:
 	trace_pnfs_mds_fallback_pg_init_read(pgio->pg_inode,
 			0, NFS4_MAX_UINT64, IOMODE_READ,
@@ -1820,7 +1836,7 @@ ff_layout_read_pagelist(struct nfs_pgio_header *hdr)
 	hdr->mds_offset = offset;
 
 	/* Start IO accounting for local read */
-	localio = ff_local_open_fh(ds->ds_clp, ds_cred, fh, FMODE_READ);
+	localio = ff_local_open_fh(lseg, idx, ds->ds_clp, ds_cred, fh, FMODE_READ);
 	if (localio) {
 		hdr->task.tk_start = ktime_get();
 		ff_layout_read_record_layoutstats_start(&hdr->task, hdr);
@@ -1896,7 +1912,7 @@ ff_layout_write_pagelist(struct nfs_pgio_header *hdr, int sync)
 	hdr->args.offset = offset;
 
 	/* Start IO accounting for local write */
-	localio = ff_local_open_fh(ds->ds_clp, ds_cred, fh,
+	localio = ff_local_open_fh(lseg, idx, ds->ds_clp, ds_cred, fh,
 				   FMODE_READ|FMODE_WRITE);
 	if (localio) {
 		hdr->task.tk_start = ktime_get();
@@ -1981,7 +1997,7 @@ static int ff_layout_initiate_commit(struct nfs_commit_data *data, int how)
 		data->args.fh = fh;
 
 	/* Start IO accounting for local commit */
-	localio = ff_local_open_fh(ds->ds_clp, ds_cred, fh,
+	localio = ff_local_open_fh(lseg, idx, ds->ds_clp, ds_cred, fh,
 				   FMODE_READ|FMODE_WRITE);
 	if (localio) {
 		data->task.tk_start = ktime_get();

fs/nfs/flexfilelayout/flexfilelayout.h

@@ -83,6 +83,7 @@ struct nfs4_ff_layout_mirror {
 	nfs4_stateid			stateid;
 	const struct cred __rcu		*ro_cred;
 	const struct cred __rcu		*rw_cred;
+	struct nfs_file_localio		nfl;
 	refcount_t			ref;
 	spinlock_t			lock;
 	unsigned long			flags;

fs/nfs/inode.c

@@ -1137,6 +1137,8 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry,
 	ctx->lock_context.open_context = ctx;
 	INIT_LIST_HEAD(&ctx->list);
 	ctx->mdsthreshold = NULL;
+	nfs_localio_file_init(&ctx->nfl);
+
 	return ctx;
 }
 EXPORT_SYMBOL_GPL(alloc_nfs_open_context);
@@ -1168,6 +1170,7 @@ static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync)
 	nfs_sb_deactive(sb);
 	put_rpccred(rcu_dereference_protected(ctx->ll_cred, 1));
 	kfree(ctx->mdsthreshold);
+	nfs_close_local_fh(&ctx->nfl);
 	kfree_rcu(ctx, rcu_head);
 }
 

fs/nfs/internal.h

@@ -455,11 +455,13 @@ extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode);
 
 #if IS_ENABLED(CONFIG_NFS_LOCALIO)
 /* localio.c */
-extern void nfs_local_disable(struct nfs_client *);
 extern void nfs_local_probe(struct nfs_client *);
+extern void nfs_local_probe_async(struct nfs_client *);
+extern void nfs_local_probe_async_work(struct work_struct *);
 extern struct nfsd_file *nfs_local_open_fh(struct nfs_client *,
 					   const struct cred *,
 					   struct nfs_fh *,
+					   struct nfs_file_localio *,
 					   const fmode_t);
 extern int nfs_local_doio(struct nfs_client *,
 			  struct nfsd_file *,
@@ -471,11 +473,12 @@ extern int nfs_local_commit(struct nfsd_file *,
 extern bool nfs_server_is_local(const struct nfs_client *clp);
 
 #else /* CONFIG_NFS_LOCALIO */
-static inline void nfs_local_disable(struct nfs_client *clp) {}
 static inline void nfs_local_probe(struct nfs_client *clp) {}
+static inline void nfs_local_probe_async(struct nfs_client *clp) {}
 static inline struct nfsd_file *
 nfs_local_open_fh(struct nfs_client *clp, const struct cred *cred,
-		  struct nfs_fh *fh, const fmode_t mode)
+		  struct nfs_fh *fh, struct nfs_file_localio *nfl,
+		  const fmode_t mode)
 {
 	return NULL;
 }