patch-2.1.32 linux/fs/nfsd/vfs.c

Next file: linux/fs/proc/root.c
Previous file: linux/fs/nfsd/stats.c
Back to the patch index
Back to the overall index

diff -u --recursive --new-file v2.1.31/linux/fs/nfsd/vfs.c linux/fs/nfsd/vfs.c
@@ -0,0 +1,1095 @@
+/*
+ * linux/fs/nfsd/vfs.c
+ *
+ * File operations used by nfsd. Some of these have been ripped from
+ * other parts of the kernel because they weren't in ksyms.c, others
+ * are partial duplicates with added or changed functionality.
+ *
+ * Note that several functions lock the inode upon which they want
+ * to act, most notably those that create directory entries. The
+ * unlock operation can take place either by calling fh_unlock within
+ * the function directly, or at a later time in fh_put(). So if you
+ * notice code paths that apparently fail to unlock the inode, don't
+ * worry--they have been taken care of.
+ *
+ * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#include <linux/version.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/locks.h>
+#include <linux/fs.h>
+#include <linux/major.h>
+#include <linux/ext2_fs.h>
+#include <linux/proc_fs.h>
+#include <linux/stat.h>
+#include <linux/fcntl.h>
+#include <linux/net.h>
+#include <linux/in.h>
+#include <linux/unistd.h>
+#include <linux/malloc.h>
+#include <netinet/in.h>
+
+#include <linux/sunrpc/svc.h>
+#include <linux/nfsd/nfsd.h>
+
+#if LINUX_VERSION_CODE >= 0x020100
+#include <asm/uaccess.h>
+#endif
+
+#define NFSDDBG_FACILITY		NFSDDBG_FILEOP
+
+/* Symbol not exported */
+static struct super_block *get_super(dev_t dev);
+
+/* Open mode for nfsd_open */
+#define OPEN_READ	0
+#define OPEN_WRITE	1
+
+/* Hack until we have a macro check for mandatory locks. */
+#ifndef IS_ISMNDLK
+#define IS_ISMNDLK(i)	(((i)->i_mode & (S_ISGID|S_ISVTX)) == S_ISGID)
+#endif
+
+/* Check for dir entries '.' and '..' */
+#define isdotent(n, l)	(l < 3 && n[0] == '.' && (l == 1 || n[1] == '.'))
+
+/*
+ * This is a cache of readahead params that help us choose the proper
+ * readahead strategy. Initially, we set all readahead parameters to 0
+ * and let the VFS handle things.
+ * If you increase the number of cached files very much, you'll need to
+ * add a hash table here.
+ */
+struct raparms {
+	struct raparms *	p_next;
+	unsigned int		p_count;
+	dev_t			p_dev;
+	ino_t			p_ino;
+	unsigned long		p_reada,
+				p_ramax,
+				p_raend,
+				p_ralen,
+				p_rawin;
+};
+
+#define FILECACHE_MAX		(2 * NFSD_MAXSERVS)
+static struct raparms		raparms[FILECACHE_MAX];
+static struct raparms *		raparm_cache = 0;
+
+/*
+ * Deny access to certain file systems
+ */
+static inline int
+fs_off_limits(struct super_block *sb)
+{
+	return !sb || sb->s_magic == NFS_SUPER_MAGIC
+	           || sb->s_magic == PROC_SUPER_MAGIC;
+}
+
+/*
+ * Check whether directory is a mount point
+ */
+static inline int
+nfsd_iscovered(struct inode *inode)
+{
+	return inode->i_mount != NULL;
+}
+
+/*
+ * Look up one component of a pathname.
+ */
+int
+nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name,
+					int len, struct svc_fh *resfh)
+{
+	struct svc_export	*exp;
+	struct super_block	*sb;
+	struct inode		*dirp, *inode;
+	int			perm, err, dotdot = 0;
+
+	dprintk("nfsd: nfsd_lookup(fh %x/%ld, %s)\n",
+			SVCFH_DEV(fhp), SVCFH_INO(fhp), name);
+
+	/* Obtain inode and export */
+	if ((err = fh_lookup(rqstp, fhp, S_IFDIR, MAY_NOP)) != 0)
+		return err;
+	dirp = fhp->fh_inode;
+	exp  = fhp->fh_export;
+
+	/* check permissions before traversing mount-points */
+	perm = nfsd_permission(exp, dirp, MAY_EXEC);
+
+	dotdot = (len == 2 && name[0] == '.' && name[1] == '.');
+	if (dotdot) {
+		if (dirp == current->fs->root) {
+			dirp->i_count++;
+			*resfh = *fhp;
+			return 0;
+		}
+
+		if (dirp->i_dev == exp->ex_dev && dirp->i_ino == exp->ex_ino) {
+			dirp->i_count++;
+			*resfh = *fhp;
+			return 0;
+		}
+	} else if (len == 1 && name[0] == '.') {
+		len = 0;
+	} else if (fs_off_limits(dirp->i_sb)) {
+		/* No lookups on NFS mounts and procfs */
+		return nfserr_noent;
+	} else if (nfsd_iscovered(dirp)) {
+		/* broken NFS client */
+		return nfserr_acces;
+	}
+	if (!dirp->i_op || !dirp->i_op->lookup)
+		return nfserr_notdir;
+ 	if (perm != 0)
+		return perm;
+	if (!len) {
+		dirp->i_count++;
+		*resfh = *fhp;
+		return 0;
+	}
+
+	dirp->i_count++;		/* lookup eats the dirp inode */
+	err = dirp->i_op->lookup(dirp, name, len, &inode);
+
+	if (err)
+		return nfserrno(-err);
+
+	/* Note that lookup() has already done a call to iget() so that
+	 * the inode returned never refers to an inode covered by a mount.
+	 * When this has happened, return the covered inode.
+	 */
+	if (!dotdot && (sb = inode->i_sb) && (inode == sb->s_mounted)) {
+		iput(inode);
+		inode = sb->s_covered;
+		inode->i_count++;
+	}
+
+	fh_compose(resfh, exp, inode);
+	return 0;
+}
+
+/*
+ * Set various file attributes.
+ */
+int
+nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap)
+{
+	struct inode	*inode;
+	int		accmode = MAY_SATTR;
+	int		ftype = 0;
+	int		imode;
+	int		err;
+
+	if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE))
+		accmode |= MAY_WRITE;
+	if (iap->ia_valid & ATTR_SIZE)
+		ftype = S_IFREG;
+
+	/* Get inode */
+	if ((err = fh_lookup(rqstp, fhp, ftype, accmode)) != 0)
+		return err;
+
+	fh_lock(fhp);			/* lock inode */
+	inode = fhp->fh_inode;
+
+	/* The size case is special... */
+	if ((iap->ia_valid & ATTR_SIZE) && S_ISREG(inode->i_mode)) {
+		if (iap->ia_size < inode->i_size) {
+			err = nfsd_permission(fhp->fh_export, inode, MAY_TRUNC);
+			if (err != 0)
+				return err;
+		}
+		if ((err = get_write_access(inode)) != 0)
+			return nfserrno(-err);
+		inode->i_size = iap->ia_size;
+		if (inode->i_op && inode->i_op->truncate)
+			inode->i_op->truncate(inode);
+		inode->i_dirt = 1;
+		put_write_access(inode);
+		iap->ia_valid &= ATTR_SIZE;
+		iap->ia_valid |= ATTR_MTIME;
+		iap->ia_mtime = CURRENT_TIME;
+	}
+
+	imode = inode->i_mode;
+	if (iap->ia_valid & ATTR_MODE) {
+		iap->ia_mode &= S_IALLUGO;
+		imode = iap->ia_mode |= (imode & ~S_IALLUGO);
+	}
+
+	/* Revoke setuid/setgid bit on chown/chgrp */
+	if ((iap->ia_valid & ATTR_UID) && (imode & S_ISUID)
+	 && iap->ia_uid != inode->i_uid) {
+		iap->ia_valid |= ATTR_MODE;
+		iap->ia_mode = imode &= ~S_ISUID;
+	}
+	if ((iap->ia_valid & ATTR_GID) && (imode & S_ISGID)
+	 && iap->ia_gid != inode->i_gid) {
+		iap->ia_valid |= ATTR_MODE;
+		iap->ia_mode = imode &= ~S_ISGID;
+	}
+
+	/* Change the attributes. */
+	if (iap->ia_valid) {
+		iap->ia_valid |= ATTR_CTIME;
+		iap->ia_ctime = CURRENT_TIME;
+		err = nfsd_notify_change(inode, iap);
+		if (err)
+			return nfserrno(-err);
+		if (EX_ISSYNC(fhp->fh_export))
+			nfsd_write_inode(inode);
+	}
+
+	return 0;
+}
+
+/*
+ * Open an existing file or directory.
+ * The wflag argument indicates write access.
+ */
+int
+nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
+			int wflag, struct file *filp)
+{
+	struct inode	*inode;
+	int		access, err;
+
+	access = wflag? MAY_WRITE : MAY_READ;
+	if ((err = fh_lookup(rqstp, fhp, type, access)) != 0)
+		return err;
+	inode = fhp->fh_inode;
+
+	/* Disallow access to files with the append-only bit set or
+	 * with mandatory locking enabled */
+	if (IS_APPEND(inode) || IS_ISMNDLK(inode))
+		return nfserr_perm;
+	if (!inode->i_op || !inode->i_op->default_file_ops)
+		return nfserr_perm;
+
+	if (wflag && (err = get_write_access(inode)) != 0)
+		return nfserrno(-err);
+
+	memset(filp, 0, sizeof(*filp));
+	filp->f_op    = inode->i_op->default_file_ops;
+	filp->f_count = 1;
+	filp->f_flags = wflag? O_WRONLY : O_RDONLY;
+	filp->f_mode  = wflag? FMODE_WRITE : FMODE_READ;
+	filp->f_inode = inode;
+
+	if (filp->f_op->open) {
+		err = filp->f_op->open(inode, filp);
+		if (err) {
+			if (wflag)
+				put_write_access(inode);
+			filp->f_count--;
+			return nfserrno(-err);
+		}
+	}
+
+	inode->i_count++;
+	return 0;
+}
+
+/*
+ * Close a file.
+ */
+void
+nfsd_close(struct file *filp)
+{
+	struct inode	*inode;
+
+	inode = filp->f_inode;
+	if (!inode->i_count)
+		printk(KERN_WARNING "nfsd: inode count == 0!\n");
+	if (filp->f_op && filp->f_op->release)
+		filp->f_op->release(inode, filp);
+	if (filp->f_mode & FMODE_WRITE)
+		put_write_access(inode);
+	iput(inode);
+}
+
+/*
+ * Sync a file
+ */
+void
+nfsd_sync(struct inode *inode, struct file *filp)
+{
+	filp->f_op->fsync(inode, filp);
+}
+
+/*
+ * Obtain the readahead parameters for the given file
+ */
+static inline struct raparms *
+nfsd_get_raparms(dev_t dev, ino_t ino)
+{
+	struct raparms	*ra, **rap, **frap = NULL;
+
+	for (rap = &raparm_cache; (ra = *rap); rap = &ra->p_next) {
+		if (ra->p_dev != dev || ra->p_ino != ino) {
+			if (ra->p_count == 0)
+				frap = rap;
+		} else
+			goto found;
+	}
+	if (!frap)
+		return NULL;
+	rap = frap;
+	ra = *frap;
+	memset(ra, 0, sizeof(*ra));
+found:
+	if (rap != &raparm_cache) {
+		*rap = ra->p_next;
+		ra->p_next   = raparm_cache;
+		raparm_cache = ra;
+	}
+	ra->p_count++;
+	return ra;
+}
+
+/*
+ * Read data from a file. count must contain the requested read count
+ * on entry. On return, *count contains the number of bytes actually read.
+ */
+int
+nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset, char *buf,
+						unsigned long *count)
+{
+	struct raparms	*ra;
+	struct inode	*inode;
+	struct file	file;
+	unsigned long	oldfs;
+	int		err;
+
+	if ((err = nfsd_open(rqstp, fhp, S_IFREG, OPEN_READ, &file)) != 0)
+		return err;
+	inode = file.f_inode;
+	if (!file.f_op->read) {
+		nfsd_close(&file);
+		return nfserr_perm;
+	}
+
+	/* Get readahead parameters */
+	if ((ra = nfsd_get_raparms(inode->i_dev, inode->i_ino)) != NULL) {
+		file.f_reada = ra->p_reada;
+		file.f_ramax = ra->p_ramax;
+		file.f_raend = ra->p_raend;
+		file.f_ralen = ra->p_ralen;
+		file.f_rawin = ra->p_rawin;
+	}
+	file.f_pos = offset;
+
+	oldfs = get_fs(); set_fs(KERNEL_DS);
+	err = file.f_op->read(file.f_inode, &file, buf, *count);
+	set_fs(oldfs);
+
+	/* Write back readahead params */
+	if (ra != NULL) {
+		dprintk("nfsd: raparms %ld %ld %ld %ld %ld\n",
+			file.f_reada, file.f_ramax, file.f_raend,
+			file.f_ralen, file.f_rawin);
+		ra->p_reada = file.f_reada;
+		ra->p_ramax = file.f_ramax;
+		ra->p_raend = file.f_raend;
+		ra->p_ralen = file.f_ralen;
+		ra->p_rawin = file.f_rawin;
+		ra->p_count -= 1;
+	}
+
+	nfsd_close(&file);
+
+	if (err < 0)
+		return nfserrno(-err);
+	*count = err;
+	return 0;
+}
+
+/*
+ * Write data to a file.
+ * The stable flag requests synchronous writes.
+ */
+int
+nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
+				char *buf, unsigned long cnt, int stable)
+{
+	struct svc_export	*exp;
+	struct file		file;
+	struct inode		*inode;
+	unsigned long		oldfs;
+	int			err;
+
+	if (!cnt)
+		return 0;
+	if ((err = nfsd_open(rqstp, fhp, S_IFREG, OPEN_WRITE, &file)) != 0)
+		return err;
+	if (!file.f_op->write) {
+		nfsd_close(&file);
+		return nfserr_perm;
+	}
+
+	inode = fhp->fh_inode;
+	exp   = fhp->fh_export;
+
+	/*
+	 * Request sync writes if
+	 *  -	the sync export option has been set, or
+	 *  -	the client requested O_SYNC behavior (NFSv3 feature).
+	 * When gathered writes have been configured for this volume,
+	 * flushing the data to disk is handled separately below.
+	 */
+	if ((stable || (stable = EX_ISSYNC(exp))) && !EX_WGATHER(exp))
+		file.f_flags |= O_SYNC;
+
+	fh_lock(fhp);			/* lock inode */
+	file.f_pos = offset;		/* set write offset */
+
+	/* Write the data. */
+	oldfs = get_fs(); set_fs(KERNEL_DS);
+	err = file.f_op->write(inode, &file, buf, cnt);
+	set_fs(oldfs);
+
+	/* clear setuid/setgid flag after write */
+	if (err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) {
+		struct iattr	ia;
+
+		ia.ia_valid = ATTR_MODE;
+		ia.ia_mode  = inode->i_mode & ~(S_ISUID | S_ISGID);
+		nfsd_notify_change(inode, &ia);
+	}
+
+	fh_unlock(fhp);			/* unlock inode */
+
+	if (err >= 0 && stable) {
+		static unsigned long	last_ino = 0;
+		static kdev_t		last_dev = NODEV;
+
+		/*
+		 * Gathered writes: If another process is currently
+		 * writing to the file, there's a high chance
+		 * this is another nfsd (triggered by a bulk write
+		 * from a client's biod). Rather than syncing the
+		 * file with each write request, we sleep for 10 msec.
+		 *
+		 * I don't know if this roughly approximates
+		 * C. Juszak's idea of gathered writes, but it's a
+		 * nice and simple solution (IMHO), and it seems to
+		 * work:-)
+		 */
+		if (EX_WGATHER(exp) && (inode->i_writecount > 1
+		 || (last_ino == inode->i_ino && last_dev == inode->i_dev))) {
+#if 0
+			current->timeout = jiffies + 10 * HZ / 1000;
+			interruptible_sleep_on(&inode->i_wait);
+#else
+			dprintk("nfsd: write defer %d\n", current->pid);
+			need_resched = 1;
+			current->timeout = jiffies + HZ / 100;
+			schedule();
+			dprintk("nfsd: write resume %d\n", current->pid);
+#endif
+		}
+
+		if (inode->i_dirt) {
+			dprintk("nfsd: write sync %d\n", current->pid);
+			nfsd_sync(inode, &file);
+			nfsd_write_inode(inode);
+		}
+		wake_up(&inode->i_wait);
+		last_ino = inode->i_ino;
+		last_dev = inode->i_dev;
+	}
+
+	nfsd_close(&file);
+
+	dprintk("nfsd: write complete\n");
+	return (err < 0)? nfserrno(-err) : 0;
+}
+
+/*
+ * Create a file (regular, directory, device, fifo).
+ * UNIX sockets not yet implemented.
+ */
+int
+nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
+		char *fname, int flen, struct iattr *iap,
+		int type, dev_t rdev, struct svc_fh *resfhp)
+{
+	struct inode	*dirp, *inode = NULL;
+	int		err;
+
+	if (!flen)
+		return nfserr_perm;
+
+	if (!(iap->ia_valid & ATTR_MODE))
+		iap->ia_mode = 0;
+
+	if ((err = fh_lookup(rqstp, fhp, S_IFDIR, MAY_CREATE)) != 0)
+		return err;
+
+	fh_lock(fhp);			/* lock directory */
+	dirp = fhp->fh_inode;
+	dirp->i_count++;		/* dirop eats the inode */
+
+	switch (type) {
+	case S_IFREG:
+		if (!dirp->i_op || !dirp->i_op->create)
+			return nfserr_perm;
+		err = dirp->i_op->create(dirp, fname, flen,
+					 iap->ia_mode, &inode);
+		break;
+	case S_IFDIR:
+		if (!dirp->i_op || !dirp->i_op->mkdir)
+			return nfserr_perm;
+		err = dirp->i_op->mkdir(dirp, fname, flen, iap->ia_mode);
+		break;
+	case S_IFCHR:
+	case S_IFBLK:
+	case S_IFIFO:
+		if (!dirp->i_op || !dirp->i_op->mknod)
+			return nfserr_perm;
+		err = dirp->i_op->mknod(dirp, fname, flen, iap->ia_mode, rdev);
+		break;
+	default:
+		iput(dirp);
+		err = -EACCES;
+	}
+
+	fh_unlock(fhp);
+
+	if (err < 0)
+		return nfserrno(-err);
+
+	/*
+	 * If the VFS call doesn't return the inode, look it up now.
+	 */
+	if (inode == NULL) {
+		dirp->i_count++;
+		err = dirp->i_op->lookup(dirp, fname, flen, &inode);
+		if (err < 0)
+			return -nfserrno(err);	/* Huh?! */
+	}
+
+	if (EX_ISSYNC(fhp->fh_export))
+		nfsd_write_inode(dirp);
+
+	/* Assemble the file handle for the newly created file */
+	fh_compose(resfhp, fhp->fh_export, inode);
+
+	/* Set file attributes. Mode has already been set and
+	 * setting uid/gid works only for root. Irix appears to
+	 * send along the gid when it tries to implement setgid
+	 * directories via NFS.
+	 */
+	if ((iap->ia_valid &= (ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) {
+		if ((err = nfsd_setattr(rqstp, resfhp, iap)) != 0) {
+			fh_put(resfhp);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * Truncate a file.
+ * The calling routines must make sure to update the ctime
+ * field and call notify_change.
+ */
+int
+nfsd_truncate(struct svc_rqst *rqstp, struct svc_fh *fhp, unsigned long size)
+{
+	struct inode	*inode;
+	int		err;
+
+	if ((err = fh_lookup(rqstp, fhp, S_IFREG, MAY_WRITE|MAY_TRUNC)) != 0)
+		return err;
+
+	fh_lock(fhp);			/* lock inode if not yet locked */
+	inode = fhp->fh_inode;
+
+	if ((err = get_write_access(inode)) != 0)
+		return nfserrno(-err);
+	inode->i_size = size;
+	if (inode->i_op && inode->i_op->truncate)
+		inode->i_op->truncate(inode);
+	inode->i_dirt = 1;
+	put_write_access(inode);
+
+	fh_unlock(fhp);
+
+	return 0;
+}
+
+/*
+ * Read a symlink. On entry, *lenp must contain the maximum path length that
+ * fits into the buffer. On return, it contains the true length.
+ */
+int
+nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp)
+{
+	struct inode	*inode;
+	unsigned long	oldfs;
+	int		err;
+
+	if ((err = fh_lookup(rqstp, fhp, S_IFLNK, MAY_READ)) != 0)
+		return err;
+	inode = fhp->fh_inode;
+
+	if (!inode->i_op || !inode->i_op->readlink)
+		return nfserr_io;
+
+	inode->i_count++;
+	oldfs = get_fs(); set_fs(KERNEL_DS);
+	err = inode->i_op->readlink(inode, buf, *lenp);
+	set_fs(oldfs);
+
+	if (err < 0)
+		return nfserrno(-err);
+	*lenp = err;
+
+	return 0;
+}
+
+/*
+ * Create a symlink and look up its inode
+ */
+int
+nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
+				char *fname, int flen,
+				char *path,  int plen,
+				struct svc_fh *resfhp)
+{
+	struct inode	*dirp, *inode;
+	int		err;
+
+	if (!flen || !plen)
+		return nfserr_noent;
+
+	if ((err = fh_lookup(rqstp, fhp, S_IFDIR, MAY_CREATE)) != 0)
+		return err;
+
+	dirp = fhp->fh_inode;
+	if (nfsd_iscovered(dirp))
+		return nfserr_perm;
+	if (!dirp->i_op || !dirp->i_op->symlink)
+		return nfserr_perm;
+
+	fh_lock(fhp);			/* lock inode */
+	dirp->i_count++;
+	err = dirp->i_op->symlink(dirp, fname, flen, path);
+	fh_unlock(fhp);			/* unlock inode */
+
+	if (err)
+		return nfserrno(-err);
+
+	if (EX_ISSYNC(fhp->fh_export))
+		nfsd_write_inode(dirp);
+
+	/*
+	 * Okay, now look up the inode of the new symlink.
+	 */
+	dirp->i_count++;		/* lookup eats the dirp inode */
+	err = dirp->i_op->lookup(dirp, fname, flen, &inode);
+	if (err)
+		return nfserrno(-err);
+
+	fh_compose(resfhp, fhp->fh_export, inode);
+	return 0;
+}
+
+/*
+ * Create a hardlink
+ */
+int
+nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
+				char *fname, int len, struct svc_fh *tfhp)
+{
+	struct inode	*dirp, *dest;
+	int		err;
+
+	if ((err = fh_lookup(rqstp, ffhp, S_IFDIR, MAY_CREATE) != 0) ||
+	    (err = fh_lookup(rqstp, tfhp, S_IFREG, MAY_NOP)) != 0)
+		return err;
+	dirp = ffhp->fh_inode;
+	dest = tfhp->fh_inode;
+
+	if (!len)
+		return nfserr_perm;
+	if (nfsd_iscovered(dirp))
+		return nfserr_acces;
+	if (dirp->i_dev != dest->i_dev)
+		return nfserr_acces;	/* FIXME: nxdev for NFSv3 */
+	if (IS_IMMUTABLE(dest) /* || IS_APPEND(dest) */ )
+		return nfserr_perm;
+	if (!dirp->i_op || !dirp->i_op->link)
+		return nfserr_perm;
+
+	fh_lock(ffhp);			/* lock directory inode */
+	dirp->i_count++;
+	err = dirp->i_op->link(dest, dirp, fname, len);
+	fh_unlock(ffhp);		/* unlock inode */
+
+	if (!err && EX_ISSYNC(ffhp->fh_export)) {
+		nfsd_write_inode(dirp);
+		nfsd_write_inode(dest);
+	}
+
+	return err? nfserrno(-err) : 0;
+}
+
+/*
+ * Rename a file
+ */
+int
+nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
+			    struct svc_fh *tfhp, char *tname, int tlen)
+{
+	struct inode	*fdir, *tdir;
+	int		err;
+
+	if ((err = fh_lookup(rqstp, ffhp, S_IFDIR, MAY_REMOVE) != 0)
+	 || (err = fh_lookup(rqstp, tfhp, S_IFDIR, MAY_CREATE)) != 0)
+		return err;
+	fdir = ffhp->fh_inode;
+	tdir = tfhp->fh_inode;
+
+	if (!flen || (fname[0] == '.' && 
+	    (flen == 1 || (flen == 2 && fname[1] == '.'))) ||
+	    !tlen || (tname[0] == '.' && 
+	    (tlen == 1 || (tlen == 2 && tname[1] == '.'))))
+		return nfserr_perm;
+
+	if (fdir->i_dev != tdir->i_dev)
+		return nfserr_acces;	/* nfserr_nxdev */
+	if (!fdir->i_op || !fdir->i_op->rename)
+		return nfserr_perm;
+
+	fh_lock(tfhp);			/* lock destination directory */
+	tdir->i_count++;
+	fdir->i_count++;
+	err = fdir->i_op->rename(fdir, fname, flen, tdir, tname, tlen, 0);
+	fh_unlock(tfhp);		/* unlock inode */
+
+	if (!err && EX_ISSYNC(tfhp->fh_export)) {
+		nfsd_write_inode(fdir);
+		nfsd_write_inode(tdir);
+	}
+
+	return err? nfserrno(-err) : 0;
+}
+
+/*
+ * Unlink a file or directory
+ */
+int
+nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
+				char *fname, int flen)
+{
+	struct inode	*dirp;
+	int		err;
+
+	if (!flen || isdotent(fname, flen))
+		return nfserr_acces;
+
+	if ((err = fh_lookup(rqstp, fhp, S_IFDIR, MAY_REMOVE)) != 0)
+		return err;
+
+	fh_lock(fhp);			/* lock inode */
+	dirp = fhp->fh_inode;
+
+	if (type == S_IFDIR) {
+		if (!dirp->i_op || !dirp->i_op->rmdir)
+			return nfserr_notdir;
+		dirp->i_count++;
+		err = dirp->i_op->rmdir(dirp, fname, flen);
+	} else {	/* other than S_IFDIR */
+		if (!dirp->i_op || !dirp->i_op->unlink)
+			return nfserr_perm;
+		dirp->i_count++;
+		err = dirp->i_op->unlink(dirp, fname, flen);
+	}
+
+	fh_unlock(fhp);			/* unlock inode */
+	if (!err && EX_ISSYNC(fhp->fh_export))
+		nfsd_write_inode(dirp);
+
+	return err? nfserrno(-err) : 0;
+}
+
+/*
+ * Read entries from a directory.
+ */
+int
+nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset, 
+			encode_dent_fn func, u32 *buffer, int *countp)
+{
+	struct readdir_cd cd;
+	struct file	file;
+	u32		*p;
+	int		oldlen, eof, err;
+
+	if (offset > ~(u32) 0)
+		return 0;
+
+	if ((err = nfsd_open(rqstp, fhp, S_IFDIR, OPEN_READ, &file)) != 0)
+		return nfserrno(-err);
+
+	if (!file.f_op->readdir) {
+		nfsd_close(&file);
+		return nfserr_notdir;
+	}
+	file.f_pos = offset;
+
+	/* Set up the readdir context */
+	memset(&cd, 0, sizeof(cd));
+	cd.rqstp  = rqstp;
+	cd.buffer = buffer;
+	cd.buflen = *countp >> 2;
+
+	/*
+	 * Read the directory entries. This silly loop is necessary because
+	 * readdir() is not guaranteed to fill up the entire buffer, but
+	 * may choose to do less.
+	 */
+	do {
+		oldlen = cd.buflen;
+
+		/*
+		dprintk("nfsd: f_op->readdir(%x/%ld @ %d) buflen = %d (%d)\n",
+			file.f_inode->i_dev, file.f_inode->i_ino,
+			(int) file.f_pos, (int) oldlen, (int) cd.buflen);
+		 */
+		err = file.f_op->readdir(file.f_inode, &file,
+					 &cd, (filldir_t) func);
+
+		if (err < 0) {
+			nfsd_close(&file);
+			return nfserrno(-err);
+		}
+		if (oldlen == cd.buflen)
+			break;
+	} while (oldlen != cd.buflen && !cd.eob);
+
+	/* If we didn't fill the buffer completely, we're at EOF */
+	eof = !cd.eob;
+
+	/* Hewlett Packard ignores the eof flag on READDIR. Some
+	 * fs-specific readdir implementations seem to reset f_pos to 0
+	 * at EOF however, causing an endless loop. */
+	if (cd.offset && !eof)
+		*cd.offset = htonl(file.f_pos);
+
+	/* Close the file */
+	nfsd_close(&file);
+
+	p = cd.buffer;
+	*p++ = 0;			/* no more entries */
+	*p++ = htonl(eof);		/* end of directory */
+	*countp = (caddr_t) p - (caddr_t) buffer;
+
+	dprintk("nfsd: readdir result %d bytes, eof %d offset %ld\n",
+				*countp, eof,
+				cd.offset? ntohl(*cd.offset) : -1);
+	return 0;
+}
+
+/*
+ * Get file system stats
+ */
+int
+nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct statfs *stat)
+{
+	struct inode		*inode;
+	struct super_block	*sb;
+	unsigned long		oldfs;
+	int			err;
+
+	if ((err = fh_lookup(rqstp, fhp, 0, MAY_NOP)) != 0)
+		return err;
+	inode = fhp->fh_inode;
+
+	if (!(sb = inode->i_sb) || !sb->s_op->statfs)
+		return nfserr_io;
+
+	oldfs = get_fs();
+	set_fs (KERNEL_DS);
+	sb->s_op->statfs(sb, stat, sizeof(*stat));
+	set_fs (oldfs);
+
+	return 0;
+}
+
+/*
+ * Check for a user's access permissions to this inode.
+ */
+int
+nfsd_permission(struct svc_export *exp, struct inode *inode, int acc)
+{
+	int		err;
+
+	if (acc == MAY_NOP)
+		return 0;
+
+	/*
+	dprintk("nfsd: permission 0x%x%s%s%s%s%s mode 0%o%s%s%s\n",
+		acc,
+		(acc & MAY_READ)?	" read"  : "",
+		(acc & MAY_WRITE)?	" write" : "",
+		(acc & MAY_EXEC)?	" exec"  : "",
+		(acc & MAY_SATTR)?	" sattr" : "",
+		(acc & MAY_TRUNC)?	" trunc" : "",
+		inode->i_mode,
+		IS_IMMUTABLE(inode)?	" immut" : "",
+		IS_APPEND(inode)?	" append" : "",
+		IS_RDONLY(inode)?	" ro" : "");
+	dprintk("      owner %d/%d user %d/%d\n",
+		inode->i_uid, inode->i_gid, current->fsuid, current->fsgid);
+	 */
+
+	if (acc & (MAY_WRITE | MAY_SATTR | MAY_TRUNC)) {
+		if (EX_RDONLY(exp) || IS_RDONLY(inode))
+			return nfserr_rofs;
+		if (S_ISDIR(inode->i_mode) && nfsd_iscovered(inode))
+			return nfserr_perm;
+		if (/* (acc & MAY_WRITE) && */ IS_IMMUTABLE(inode))
+			return nfserr_perm;
+	}
+	if ((acc & MAY_TRUNC) && IS_APPEND(inode))
+		return nfserr_perm;
+
+	/*
+	 * The file owner always gets access permission. This is to make
+	 * file access work even when the client has done a fchmod(fd, 0).
+	 *
+	 * However, `cp foo bar' should fail nevertheless when bar is
+	 * readonly. A sensible way to do this might be to reject all
+	 * attempts to truncate a read-only file, because a creat() call
+	 * always implies file truncation.
+	 */
+	if (inode->i_uid == current->fsuid /* && !(acc & MAY_TRUNC) */)
+		return 0;
+
+	err = permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC));
+
+	/* Allow read access to binaries even when mode 111 */
+	if (err == -EPERM && S_ISREG(inode->i_mode) && acc == MAY_READ)
+		err = permission(inode, MAY_EXEC);
+
+	return err? nfserrno(-err) : 0;
+}
+
+/*
+ * Look up the inode for a given FH.
+ */
+struct inode *
+nfsd_iget(dev_t dev, ino_t ino)
+{
+	struct super_block	*sb;
+
+	if (!(sb = get_super(dev)) || fs_off_limits(sb))
+		return NULL;
+	return __iget(sb, ino, 0);
+}
+
+/*
+ * Write the inode if dirty (copy of fs/inode.c:write_inode)
+ */
+void
+nfsd_write_inode(struct inode *inode)
+{
+	struct super_block	*sb = inode->i_sb;
+
+	if (!inode->i_dirt)
+		return;
+	while (inode->i_lock) {
+		sleep_on(&inode->i_wait);
+		if (!inode->i_dirt)
+			return;
+	}
+	if (!sb || !sb->s_op || !sb->s_op->write_inode) {
+		inode->i_dirt = 0;
+		return;
+	}
+	inode->i_lock = 1;
+	sb->s_op->write_inode(inode);
+	inode->i_lock = 0;
+	wake_up(&inode->i_wait);
+}
+
+/*
+ * Look up the root inode of the parent fs.
+ * We have to go through iget in order to allow for wait_on_inode.
+ */
+int
+nfsd_parentdev(dev_t* devp)
+{
+	struct super_block	*sb;
+
+	if (!(sb = get_super(*devp)) || !sb->s_covered)
+		return 0;
+	if (*devp == sb->s_covered->i_dev)
+		return 0;
+	*devp = sb->s_covered->i_dev;
+	return 1;
+}
+
+/* Duplicated here from fs/super.c because it's not exported */
+static struct super_block *
+get_super(dev_t dev)
+{
+	struct super_block *s;
+
+	if (!dev)
+		return NULL;
+	s = 0 + super_blocks;
+	while (s < NR_SUPER + super_blocks)
+		if (s->s_dev == dev) {
+			wait_on_super(s);
+			if (s->s_dev == dev)
+				return s;
+			s = 0 + super_blocks;
+		} else
+			s++;
+	return NULL;
+}
+
+/*
+ * This is a copy from fs/inode.c because it wasn't exported.
+ */
+int
+nfsd_notify_change(struct inode *inode, struct iattr *attr)
+{
+	int retval;
+
+	if (inode->i_sb && inode->i_sb->s_op &&
+	    inode->i_sb->s_op->notify_change)
+		return inode->i_sb->s_op->notify_change(inode, attr);
+
+	if ((retval = inode_change_ok(inode, attr)) != 0)
+		return retval;
+
+	inode_setattr(inode, attr);
+	return 0;
+}
+
+/*
+ * Initialize readahead param cache
+ */
+void
+nfsd_racache_init(void)
+{
+	int	i;
+
+	if (raparm_cache)
+		return;
+	memset(raparms, 0, sizeof(raparms));
+	for (i = 0; i < FILECACHE_MAX - 1; i++) {
+		raparms[i].p_next = raparms + i + 1;
+	}
+	raparm_cache = raparms;
+}

FUNET's LINUX-ADM group, linux-adm@nic.funet.fi
TCL-scripts by Sam Shen, slshen@lbl.gov