1 files changed, 169 insertions, 78 deletions

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 29cb7b812d71..cd689df2ca5d 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -31,6 +31,7 @@
 #include <linux/exportfs.h>
 #include <linux/writeback.h>
 #include <linux/security.h>
+#include <linux/sunrpc/xdr.h>
 
 #include "xdr3.h"
 
@@ -47,6 +48,8 @@
 
 #define NFSDDBG_FACILITY		NFSDDBG_FILEOP
 
+bool nfsd_disable_splice_read __read_mostly;
+
 /**
  * nfserrno - Map Linux errnos to NFS errnos
  * @errno: POSIX(-ish) error code to be mapped
@@ -71,7 +74,6 @@ nfserrno (int errno)
 		{ nfserr_acces, -EACCES },
 		{ nfserr_exist, -EEXIST },
 		{ nfserr_xdev, -EXDEV },
-		{ nfserr_mlink, -EMLINK },
 		{ nfserr_nodev, -ENODEV },
 		{ nfserr_notdir, -ENOTDIR },
 		{ nfserr_isdir, -EISDIR },
@@ -245,7 +247,7 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	struct dentry		*dentry;
 	int			host_err;
 
-	dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name);
+	trace_nfsd_vfs_lookup(rqstp, fhp, name, len);
 
 	dparent = fhp->fh_dentry;
 	exp = exp_get(fhp->fh_export);
@@ -265,7 +267,8 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
 				goto out_nfserr;
 		}
 	} else {
-		dentry = lookup_one_len_unlocked(name, dparent, len);
+		dentry = lookup_one_unlocked(&nop_mnt_idmap,
+					     &QSTR_LEN(name, len), dparent);
 		host_err = PTR_ERR(dentry);
 		if (IS_ERR(dentry))
 			goto out_nfserr;
@@ -500,6 +503,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	bool		size_change = (iap->ia_valid & ATTR_SIZE);
 	int		retries;
 
+	trace_nfsd_vfs_setattr(rqstp, fhp, iap, guardtime);
+
 	if (iap->ia_valid & ATTR_SIZE) {
 		accmode |= NFSD_MAY_WRITE|NFSD_MAY_OWNER_OVERRIDE;
 		ftype = S_IFREG;
@@ -923,7 +928,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
 	 * directories, but we never have and it doesn't seem to have
 	 * caused anyone a problem.  If we were to change this, note
 	 * also that our filldir callbacks would need a variant of
-	 * lookup_one_len that doesn't check permissions.
+	 * lookup_one_positive_unlocked() that doesn't check permissions.
 	 */
 	if (type == S_IFREG)
 		may_flags |= NFSD_MAY_OWNER_OVERRIDE;
@@ -1082,23 +1087,23 @@ __be32 nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	unsigned long v, total;
 	struct iov_iter iter;
 	loff_t ppos = offset;
-	struct page *page;
 	ssize_t host_err;
+	size_t len;
 
 	v = 0;
 	total = *count;
 	while (total) {
-		page = *(rqstp->rq_next_page++);
-		rqstp->rq_vec[v].iov_base = page_address(page) + base;
-		rqstp->rq_vec[v].iov_len = min_t(size_t, total, PAGE_SIZE - base);
-		total -= rqstp->rq_vec[v].iov_len;
+		len = min_t(size_t, total, PAGE_SIZE - base);
+		bvec_set_page(&rqstp->rq_bvec[v], *(rqstp->rq_next_page++),
+			      len, base);
+		total -= len;
 		++v;
 		base = 0;
 	}
-	WARN_ON_ONCE(v > ARRAY_SIZE(rqstp->rq_vec));
+	WARN_ON_ONCE(v > rqstp->rq_maxpages);
 
 	trace_nfsd_read_vector(rqstp, fhp, offset, *count);
-	iov_iter_kvec(&iter, ITER_DEST, rqstp->rq_vec, v, *count);
+	iov_iter_bvec(&iter, ITER_DEST, rqstp->rq_bvec, v, *count);
 	host_err = vfs_iter_read(file, &iter, &ppos, 0);
 	return nfsd_finish_read(rqstp, fhp, file, offset, count, eof, host_err);
 }
@@ -1140,11 +1145,27 @@ static int wait_for_concurrent_writes(struct file *file)
 	return err;
 }
 
+/**
+ * nfsd_vfs_write - write data to an already-open file
+ * @rqstp: RPC execution context
+ * @fhp: File handle of file to write into
+ * @nf: An open file matching @fhp
+ * @offset: Byte offset of start
+ * @payload: xdr_buf containing the write payload
+ * @cnt: IN: number of bytes to write, OUT: number of bytes actually written
+ * @stable: An NFS stable_how value
+ * @verf: NFS WRITE verifier
+ *
+ * Upon return, caller must invoke fh_put on @fhp.
+ *
+ * Return values:
+ *   An nfsstat value in network byte order.
+ */
 __be32
-nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf,
-				loff_t offset, struct kvec *vec, int vlen,
-				unsigned long *cnt, int stable,
-				__be32 *verf)
+nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
+	       struct nfsd_file *nf, loff_t offset,
+	       const struct xdr_buf *payload, unsigned long *cnt,
+	       int stable, __be32 *verf)
 {
 	struct nfsd_net		*nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
 	struct file		*file = nf->nf_file;
@@ -1159,6 +1180,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf,
 	unsigned int		pflags = current->flags;
 	rwf_t			flags = 0;
 	bool			restore_flags = false;
+	unsigned int		nvecs;
 
 	trace_nfsd_write_opened(rqstp, fhp, offset, *cnt);
 
@@ -1186,7 +1208,8 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf,
 	if (stable && !fhp->fh_use_wgather)
 		flags |= RWF_SYNC;
 
-	iov_iter_kvec(&iter, ITER_SOURCE, vec, vlen, *cnt);
+	nvecs = xdr_buf_to_bvec(rqstp->rq_bvec, rqstp->rq_maxpages, payload);
+	iov_iter_bvec(&iter, ITER_SOURCE, rqstp->rq_bvec, nvecs, *cnt);
 	since = READ_ONCE(file->f_wb_err);
 	if (verf)
 		nfsd_copy_write_verifier(verf, nn);
@@ -1237,6 +1260,8 @@ out_nfserr:
  */
 bool nfsd_read_splice_ok(struct svc_rqst *rqstp)
 {
+	if (nfsd_disable_splice_read)
+		return false;
 	switch (svc_auth_flavor(rqstp)) {
 	case RPC_AUTH_GSS_KRB5I:
 	case RPC_AUTH_GSS_KRB5P:
@@ -1284,14 +1309,24 @@ __be32 nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	return err;
 }
 
-/*
- * Write data to a file.
- * The stable flag requests synchronous writes.
- * N.B. After this call fhp needs an fh_put
+/**
+ * nfsd_write - open a file and write data to it
+ * @rqstp: RPC execution context
+ * @fhp: File handle of file to write into; nfsd_write() may modify it
+ * @offset: Byte offset of start
+ * @payload: xdr_buf containing the write payload
+ * @cnt: IN: number of bytes to write, OUT: number of bytes actually written
+ * @stable: An NFS stable_how value
+ * @verf: NFS WRITE verifier
+ *
+ * Upon return, caller must invoke fh_put on @fhp.
+ *
+ * Return values:
+ *   An nfsstat value in network byte order.
  */
 __be32
 nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
-	   struct kvec *vec, int vlen, unsigned long *cnt, int stable,
+	   const struct xdr_buf *payload, unsigned long *cnt, int stable,
 	   __be32 *verf)
 {
 	struct nfsd_file *nf;
@@ -1303,8 +1338,8 @@ nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
 	if (err)
 		goto out;
 
-	err = nfsd_vfs_write(rqstp, fhp, nf, offset, vec,
-			vlen, cnt, stable, verf);
+	err = nfsd_vfs_write(rqstp, fhp, nf, offset, payload, cnt,
+			     stable, verf);
 	nfsd_file_put(nf);
 out:
 	trace_nfsd_write_done(rqstp, fhp, offset, *cnt);
@@ -1340,6 +1375,8 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf,
 	loff_t			start, end;
 	struct nfsd_net		*nn;
 
+	trace_nfsd_commit_start(rqstp, fhp, offset, count);
+
 	/*
 	 * Convert the client-provided (offset, count) range to a
 	 * (start, end) range. If the client-provided range falls
@@ -1378,6 +1415,7 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf,
 	} else
 		nfsd_copy_write_verifier(verf, nn);
 
+	trace_nfsd_commit_done(rqstp, fhp, offset, count);
 	return err;
 }
 
@@ -1461,7 +1499,7 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	struct inode	*dirp;
 	struct iattr	*iap = attrs->na_iattr;
 	__be32		err;
-	int		host_err;
+	int		host_err = 0;
 
 	dentry = fhp->fh_dentry;
 	dirp = d_inode(dentry);
@@ -1488,28 +1526,15 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp,
 			nfsd_check_ignore_resizing(iap);
 		break;
 	case S_IFDIR:
-		host_err = vfs_mkdir(&nop_mnt_idmap, dirp, dchild, iap->ia_mode);
-		if (!host_err && unlikely(d_unhashed(dchild))) {
-			struct dentry *d;
-			d = lookup_one_len(dchild->d_name.name,
-					   dchild->d_parent,
-					   dchild->d_name.len);
-			if (IS_ERR(d)) {
-				host_err = PTR_ERR(d);
-				break;
-			}
-			if (unlikely(d_is_negative(d))) {
-				dput(d);
-				err = nfserr_serverfault;
-				goto out;
-			}
+		dchild = vfs_mkdir(&nop_mnt_idmap, dirp, dchild, iap->ia_mode);
+		if (IS_ERR(dchild)) {
+			host_err = PTR_ERR(dchild);
+		} else if (d_is_negative(dchild)) {
+			err = nfserr_serverfault;
+			goto out;
+		} else if (unlikely(dchild != resfhp->fh_dentry)) {
 			dput(resfhp->fh_dentry);
-			resfhp->fh_dentry = dget(d);
-			err = fh_update(resfhp);
-			dput(dchild);
-			dchild = d;
-			if (err)
-				goto out;
+			resfhp->fh_dentry = dget(dchild);
 		}
 		break;
 	case S_IFCHR:
@@ -1530,7 +1555,8 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	err = nfsd_create_setattr(rqstp, fhp, resfhp, attrs);
 
 out:
-	dput(dchild);
+	if (!IS_ERR(dchild))
+		dput(dchild);
 	return err;
 
 out_nfserr:
@@ -1553,6 +1579,8 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	__be32		err;
 	int		host_err;
 
+	trace_nfsd_vfs_create(rqstp, fhp, type, fname, flen);
+
 	if (isdotent(fname, flen))
 		return nfserr_exist;
 
@@ -1567,7 +1595,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		return nfserrno(host_err);
 
 	inode_lock_nested(dentry->d_inode, I_MUTEX_PARENT);
-	dchild = lookup_one_len(fname, dentry, flen);
+	dchild = lookup_one(&nop_mnt_idmap, &QSTR_LEN(fname, flen), dentry);
 	host_err = PTR_ERR(dchild);
 	if (IS_ERR(dchild)) {
 		err = nfserrno(host_err);
@@ -1653,6 +1681,8 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	__be32		err, cerr;
 	int		host_err;
 
+	trace_nfsd_vfs_symlink(rqstp, fhp, fname, flen, path);
+
 	err = nfserr_noent;
 	if (!flen || path[0] == '\0')
 		goto out;
@@ -1672,7 +1702,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
 
 	dentry = fhp->fh_dentry;
 	inode_lock_nested(dentry->d_inode, I_MUTEX_PARENT);
-	dnew = lookup_one_len(fname, dentry, flen);
+	dnew = lookup_one(&nop_mnt_idmap, &QSTR_LEN(fname, flen), dentry);
 	if (IS_ERR(dnew)) {
 		err = nfserrno(PTR_ERR(dnew));
 		inode_unlock(dentry->d_inode);
@@ -1699,9 +1729,17 @@ out:
 	return err;
 }
 
-/*
- * Create a hardlink
- * N.B. After this call _both_ ffhp and tfhp need an fh_put
+/**
+ * nfsd_link - create a link
+ * @rqstp: RPC transaction context
+ * @ffhp: the file handle of the directory where the new link is to be created
+ * @name: the filename of the new link
+ * @len: the length of @name in octets
+ * @tfhp: the file handle of an existing file object
+ *
+ * After this call _both_ ffhp and tfhp need an fh_put.
+ *
+ * Returns a generic NFS status code in network byte-order.
  */
 __be32
 nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
@@ -1709,9 +1747,12 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
 {
 	struct dentry	*ddir, *dnew, *dold;
 	struct inode	*dirp;
+	int		type;
 	__be32		err;
 	int		host_err;
 
+	trace_nfsd_vfs_link(rqstp, ffhp, tfhp, name, len);
+
 	err = fh_verify(rqstp, ffhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
@@ -1728,19 +1769,19 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
 	if (isdotent(name, len))
 		goto out;
 
+	err = nfs_ok;
+	type = d_inode(tfhp->fh_dentry)->i_mode & S_IFMT;
 	host_err = fh_want_write(tfhp);
-	if (host_err) {
-		err = nfserrno(host_err);
+	if (host_err)
 		goto out;
-	}
 
 	ddir = ffhp->fh_dentry;
 	dirp = d_inode(ddir);
 	inode_lock_nested(dirp, I_MUTEX_PARENT);
 
-	dnew = lookup_one_len(name, ddir, len);
+	dnew = lookup_one(&nop_mnt_idmap, &QSTR_LEN(name, len), ddir);
 	if (IS_ERR(dnew)) {
-		err = nfserrno(PTR_ERR(dnew));
+		host_err = PTR_ERR(dnew);
 		goto out_unlock;
 	}
 
@@ -1756,17 +1797,26 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
 	fh_fill_post_attrs(ffhp);
 	inode_unlock(dirp);
 	if (!host_err) {
-		err = nfserrno(commit_metadata(ffhp));
-		if (!err)
-			err = nfserrno(commit_metadata(tfhp));
-	} else {
-		err = nfserrno(host_err);
+		host_err = commit_metadata(ffhp);
+		if (!host_err)
+			host_err = commit_metadata(tfhp);
 	}
+
 	dput(dnew);
 out_drop_write:
 	fh_drop_write(tfhp);
+	if (host_err == -EBUSY) {
+		/*
+		 * See RFC 8881 Section 18.9.4 para 1-2: NFSv4 LINK
+		 * wants a status unique to the object type.
+		 */
+		if (type != S_IFDIR)
+			err = nfserr_file_open;
+		else
+			err = nfserr_acces;
+	}
 out:
-	return err;
+	return err != nfs_ok ? err : nfserrno(host_err);
 
 out_dput:
 	dput(dnew);
@@ -1795,9 +1845,19 @@ nfsd_has_cached_files(struct dentry *dentry)
 	return ret;
 }
 
-/*
- * Rename a file
- * N.B. After this call _both_ ffhp and tfhp need an fh_put
+/**
+ * nfsd_rename - rename a directory entry
+ * @rqstp: RPC transaction context
+ * @ffhp: the file handle of parent directory containing the entry to be renamed
+ * @fname: the filename of directory entry to be renamed
+ * @flen: the length of @fname in octets
+ * @tfhp: the file handle of parent directory to contain the renamed entry
+ * @tname: the filename of the new entry
+ * @tlen: the length of @tlen in octets
+ *
+ * After this call _both_ ffhp and tfhp need an fh_put.
+ *
+ * Returns a generic NFS status code in network byte-order.
  */
 __be32
 nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
@@ -1805,10 +1865,13 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
 {
 	struct dentry	*fdentry, *tdentry, *odentry, *ndentry, *trap;
 	struct inode	*fdir, *tdir;
+	int		type = S_IFDIR;
 	__be32		err;
 	int		host_err;
 	bool		close_cached = false;
 
+	trace_nfsd_vfs_rename(rqstp, ffhp, tfhp, fname, flen, tname, tlen);
+
 	err = fh_verify(rqstp, ffhp, S_IFDIR, NFSD_MAY_REMOVE);
 	if (err)
 		goto out;
@@ -1851,7 +1914,7 @@ retry:
 	if (err != nfs_ok)
 		goto out_unlock;
 
-	odentry = lookup_one_len(fname, fdentry, flen);
+	odentry = lookup_one(&nop_mnt_idmap, &QSTR_LEN(fname, flen), fdentry);
 	host_err = PTR_ERR(odentry);
 	if (IS_ERR(odentry))
 		goto out_nfserr;
@@ -1862,11 +1925,14 @@ retry:
 	host_err = -EINVAL;
 	if (odentry == trap)
 		goto out_dput_old;
+	type = d_inode(odentry)->i_mode & S_IFMT;
 
-	ndentry = lookup_one_len(tname, tdentry, tlen);
+	ndentry = lookup_one(&nop_mnt_idmap, &QSTR_LEN(tname, tlen), tdentry);
 	host_err = PTR_ERR(ndentry);
 	if (IS_ERR(ndentry))
 		goto out_dput_old;
+	if (d_inode(ndentry))
+		type = d_inode(ndentry)->i_mode & S_IFMT;
 	host_err = -ENOTEMPTY;
 	if (ndentry == trap)
 		goto out_dput_new;
@@ -1904,7 +1970,18 @@ retry:
  out_dput_old:
 	dput(odentry);
  out_nfserr:
-	err = nfserrno(host_err);
+	if (host_err == -EBUSY) {
+		/*
+		 * See RFC 8881 Section 18.26.4 para 1-3: NFSv4 RENAME
+		 * wants a status unique to the object type.
+		 */
+		if (type != S_IFDIR)
+			err = nfserr_file_open;
+		else
+			err = nfserr_acces;
+	} else {
+		err = nfserrno(host_err);
+	}
 
 	if (!close_cached) {
 		fh_fill_post_attrs(ffhp);
@@ -1931,9 +2008,17 @@ out:
 	return err;
 }
 
-/*
- * Unlink a file or directory
- * N.B. After this call fhp needs an fh_put
+/**
+ * nfsd_unlink - remove a directory entry
+ * @rqstp: RPC transaction context
+ * @fhp: the file handle of the parent directory to be modified
+ * @type: enforced file type of the object to be removed
+ * @fname: the name of directory entry to be removed
+ * @flen: length of @fname in octets
+ *
+ * After this call fhp needs an fh_put.
+ *
+ * Returns a generic NFS status code in network byte-order.
  */
 __be32
 nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
@@ -1945,6 +2030,8 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	__be32		err;
 	int		host_err;
 
+	trace_nfsd_vfs_unlink(rqstp, fhp, fname, flen);
+
 	err = nfserr_acces;
 	if (!flen || isdotent(fname, flen))
 		goto out;
@@ -1960,7 +2047,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	dirp = d_inode(dentry);
 	inode_lock_nested(dirp, I_MUTEX_PARENT);
 
-	rdentry = lookup_one_len(fname, dentry, flen);
+	rdentry = lookup_one(&nop_mnt_idmap, &QSTR_LEN(fname, flen), dentry);
 	host_err = PTR_ERR(rdentry);
 	if (IS_ERR(rdentry))
 		goto out_unlock;
@@ -2007,15 +2094,17 @@ out_drop_write:
 	fh_drop_write(fhp);
 out_nfserr:
 	if (host_err == -EBUSY) {
-		/* name is mounted-on. There is no perfect
-		 * error status.
+		/*
+		 * See RFC 8881 Section 18.25.4 para 4: NFSv4 REMOVE
+		 * wants a status unique to the object type.
 		 */
-		err = nfserr_file_open;
-	} else {
-		err = nfserrno(host_err);
+		if (type != S_IFDIR)
+			err = nfserr_file_open;
+		else
+			err = nfserr_acces;
 	}
 out:
-	return err;
+	return err != nfs_ok ? err : nfserrno(host_err);
 out_unlock:
 	inode_unlock(dirp);
 	goto out_drop_write;
@@ -2231,6 +2320,8 @@ nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat, in
 {
 	__be32 err;
 
+	trace_nfsd_vfs_statfs(rqstp, fhp);
+
 	err = fh_verify(rqstp, fhp, 0, NFSD_MAY_NOP | access);
 	if (!err) {
 		struct path path = {