vfs.c

来自「Linux Kernel 2.6.9 for OMAP1710」· C语言 代码 · 共 1,815 行 · 第 1/3 页

			case nfs_ok:
				result |= map->access;
				break;
				
			/* the following error codes just mean the access was not allowed,
			 * rather than an error occurred */
			case nfserr_rofs:
			case nfserr_acces:
			case nfserr_perm:
				/* simply don't "or" in the access bit. */
				break;
			default:
				error = err2;
				goto out;
			}
		}
	}
	*access = result;
	if (supported)
		*supported = sresult;

 out:
	return error;
}
#endif /* CONFIG_NFSD_V3 */



/*
 * Open an existing file or directory.
 * The access argument indicates the type of open (read/write/lock)
 * N.B. After this call fhp needs an fh_put
 */
int
nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
			int access, struct file **filp)
{
	struct dentry	*dentry;
	struct inode	*inode;
	int		flags = O_RDONLY|O_LARGEFILE, err;

	/*
	 * If we get here, then the client has already done an "open",
	 * and (hopefully) checked permission - so allow OWNER_OVERRIDE
	 * in case a chmod has now revoked permission.
	 */
	err = fh_verify(rqstp, fhp, type, access | MAY_OWNER_OVERRIDE);
	if (err)
		goto out;

	dentry = fhp->fh_dentry;
	inode = dentry->d_inode;

	/* Disallow access to files with the append-only bit set or
	 * with mandatory locking enabled
	 */
	err = nfserr_perm;
	if (IS_APPEND(inode) || IS_ISMNDLK(inode))
		goto out;
	if (!inode->i_fop)
		goto out;

	/*
	 * Check to see if there are any leases on this file.
	 * This may block while leases are broken.
	 */
	err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0));
	if (err) /* NOMEM or WOULDBLOCK */
		goto out_nfserr;

	if (access & MAY_WRITE) {
		flags = O_WRONLY|O_LARGEFILE;

		DQUOT_INIT(inode);
	}
	*filp = dentry_open(dget(dentry), mntget(fhp->fh_export->ex_mnt), flags);
	if (IS_ERR(*filp))
		err = PTR_ERR(*filp);
out_nfserr:
	if (err)
		err = nfserrno(err);
out:
	return err;
}

/*
 * Close a file.
 */
void
nfsd_close(struct file *filp)
{
	fput(filp);
}

/*
 * Sync a file
 * As this calls fsync (not fdatasync) there is no need for a write_inode
 * after it.
 */
inline void nfsd_dosync(struct file *filp, struct dentry *dp, 
			struct file_operations *fop)
{
	struct inode *inode = dp->d_inode;
	int (*fsync) (struct file *, struct dentry *, int);

	filemap_fdatawrite(inode->i_mapping);
	if (fop && (fsync = fop->fsync))
		fsync(filp, dp, 0);
	filemap_fdatawait(inode->i_mapping);
}
	

void
nfsd_sync(struct file *filp)
{
	struct inode *inode = filp->f_dentry->d_inode;
	dprintk("nfsd: sync file %s\n", filp->f_dentry->d_name.name);
	down(&inode->i_sem);
	nfsd_dosync(filp, filp->f_dentry, filp->f_op);
	up(&inode->i_sem);
}

void
nfsd_sync_dir(struct dentry *dp)
{
	nfsd_dosync(NULL, dp, dp->d_inode->i_fop);
}

/*
 * Obtain the readahead parameters for the file
 * specified by (dev, ino).
 */
static spinlock_t ra_lock = SPIN_LOCK_UNLOCKED;

static inline struct raparms *
nfsd_get_raparms(dev_t dev, ino_t ino)
{
	struct raparms	*ra, **rap, **frap = NULL;
	int depth = 0;

	spin_lock(&ra_lock);
	for (rap = &raparm_cache; (ra = *rap); rap = &ra->p_next) {
		if (ra->p_ino == ino && ra->p_dev == dev)
			goto found;
		depth++;
		if (ra->p_count == 0)
			frap = rap;
	}
	depth = nfsdstats.ra_size*11/10;
	if (!frap) {	
		spin_unlock(&ra_lock);
		return NULL;
	}
	rap = frap;
	ra = *frap;
	ra->p_dev = dev;
	ra->p_ino = ino;
	ra->p_set = 0;
found:
	if (rap != &raparm_cache) {
		*rap = ra->p_next;
		ra->p_next   = raparm_cache;
		raparm_cache = ra;
	}
	ra->p_count++;
	nfsdstats.ra_depth[depth*10/nfsdstats.ra_size]++;
	spin_unlock(&ra_lock);
	return ra;
}

/*
 * Grab and keep cached pages assosiated with a file in the svc_rqst
 * so that they can be passed to the netowork sendmsg/sendpage routines
 * directrly. They will be released after the sending has completed.
 */
static int
nfsd_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset , unsigned long size)
{
	unsigned long count = desc->count;
	struct svc_rqst *rqstp = desc->arg.data;

	if (size > count)
		size = count;

	if (rqstp->rq_res.page_len == 0) {
		get_page(page);
		rqstp->rq_respages[rqstp->rq_resused++] = page;
		rqstp->rq_res.page_base = offset;
		rqstp->rq_res.page_len = size;
	} else if (page != rqstp->rq_respages[rqstp->rq_resused-1]) {
		get_page(page);
		rqstp->rq_respages[rqstp->rq_resused++] = page;
		rqstp->rq_res.page_len += size;
	} else {
		rqstp->rq_res.page_len += size;
	}

	desc->count = count - size;
	desc->written += size;
	return size;
}

/*
 * Read data from a file. count must contain the requested read count
 * on entry. On return, *count contains the number of bytes actually read.
 * N.B. After this call fhp needs an fh_put
 */
int
nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
          struct kvec *vec, int vlen, unsigned long *count)
{
	struct raparms	*ra;
	mm_segment_t	oldfs;
	int		err;
	struct file	*file;
	struct inode	*inode;

	err = nfsd_open(rqstp, fhp, S_IFREG, MAY_READ, &file);
	if (err)
		goto out;
	err = nfserr_perm;
	inode = file->f_dentry->d_inode;
#ifdef MSNFS
	if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) &&
		(!lock_may_read(inode, offset, *count)))
		goto out_close;
#endif

	/* Get readahead parameters */
	ra = nfsd_get_raparms(inode->i_sb->s_dev, inode->i_ino);

	if (ra && ra->p_set)
		file->f_ra = ra->p_ra;

	if (file->f_op->sendfile) {
		svc_pushback_unused_pages(rqstp);
		err = file->f_op->sendfile(file, &offset, *count,
						 nfsd_read_actor, rqstp);
	} else {
		oldfs = get_fs();
		set_fs(KERNEL_DS);
		err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset);
		set_fs(oldfs);
	}

	/* Write back readahead params */
	if (ra) {
		spin_lock(&ra_lock);
		ra->p_ra = file->f_ra;
		ra->p_set = 1;
		ra->p_count--;
		spin_unlock(&ra_lock);
	}

	if (err >= 0) {
		nfsdstats.io_read += err;
		*count = err;
		err = 0;
		dnotify_parent(file->f_dentry, DN_ACCESS);
	} else 
		err = nfserrno(err);
out_close:
	nfsd_close(file);
out:
	return err;
}

/*
 * Write data to a file.
 * The stable flag requests synchronous writes.
 * N.B. After this call fhp needs an fh_put
 */
int
nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
				struct kvec *vec, int vlen,
	   			unsigned long cnt, int *stablep)
{
	struct svc_export	*exp;
	struct file		*file;
	struct dentry		*dentry;
	struct inode		*inode;
	mm_segment_t		oldfs;
	int			err = 0;
	int			stable = *stablep;

	err = nfsd_open(rqstp, fhp, S_IFREG, MAY_WRITE, &file);
	if (err)
		goto out;
	if (!cnt)
		goto out_close;
	err = nfserr_perm;

#ifdef MSNFS
	if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) &&
		(!lock_may_write(file->f_dentry->d_inode, offset, cnt)))
		goto out_close;
#endif

	dentry = file->f_dentry;
	inode = dentry->d_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).
	 *  -   The file system doesn't support fsync().
	 * When gathered writes have been configured for this volume,
	 * flushing the data to disk is handled separately below.
	 */

	if (file->f_op->fsync == 0) {/* COMMIT3 cannot work */
	       stable = 2;
	       *stablep = 2; /* FILE_SYNC */
	}

	if (!EX_ISSYNC(exp))
		stable = 0;
	if (stable && !EX_WGATHER(exp))
		file->f_flags |= O_SYNC;

	/* Write the data. */
	oldfs = get_fs(); set_fs(KERNEL_DS);
	err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset);
	set_fs(oldfs);
	if (err >= 0) {
		nfsdstats.io_write += cnt;
		dnotify_parent(file->f_dentry, DN_MODIFY);
	}

	/* clear setuid/setgid flag after write */
	if (err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) {
		struct iattr	ia;
		ia.ia_valid = ATTR_KILL_SUID | ATTR_KILL_SGID;

		down(&inode->i_sem);
		notify_change(dentry, &ia);
		up(&inode->i_sem);
	}

	if (err >= 0 && stable) {
		static ino_t	last_ino;
		static dev_t	last_dev;

		/*
		 * 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)) {
			if (atomic_read(&inode->i_writecount) > 1
			    || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) {
				dprintk("nfsd: write defer %d\n", current->pid);
				set_current_state(TASK_UNINTERRUPTIBLE);
				schedule_timeout((HZ+99)/100);
				dprintk("nfsd: write resume %d\n", current->pid);
			}

			if (inode->i_state & I_DIRTY) {
				dprintk("nfsd: write sync %d\n", current->pid);
				nfsd_sync(file);
			}
#if 0
			wake_up(&inode->i_wait);
#endif
		}
		last_ino = inode->i_ino;
		last_dev = inode->i_sb->s_dev;
	}

	dprintk("nfsd: write complete err=%d\n", err);
	if (err >= 0)
		err = 0;
	else 
		err = nfserrno(err);
out_close:
	nfsd_close(file);
out:
	return err;
}


#ifdef CONFIG_NFSD_V3
/*
 * Commit all pending writes to stable storage.
 * Strictly speaking, we could sync just the indicated file region here,
 * but there's currently no way we can ask the VFS to do so.
 *
 * Unfortunately we cannot lock the file to make sure we return full WCC
 * data to the client, as locking happens lower down in the filesystem.
 */
int
nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp,
               loff_t offset, unsigned long count)
{
	struct file	*file;
	int		err;

	if ((u64)count > ~(u64)offset)
		return nfserr_inval;

	if ((err = nfsd_open(rqstp, fhp, S_IFREG, MAY_WRITE, &file)) != 0)
		return err;
	if (EX_ISSYNC(fhp->fh_export)) {
		if (file->f_op && file->f_op->fsync) {
			nfsd_sync(file);
		} else {
			err = nfserr_notsupp;
		}
	}

	nfsd_close(file);
	return err;
}
#endif /* CONFIG_NFSD_V3 */

/*
 * Create a file (regular, directory, device, fifo); UNIX sockets 
 * not yet implemented.
 * If the response fh has been verified, the parent directory should
 * already be locked. Note that the parent directory is left locked.
 *
 * N.B. Every call to nfsd_create needs an fh_put for _both_ fhp and resfhp
 */
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 dentry	*dentry, *dchild = NULL;
	struct inode	*dirp;
	int		err;

	err = nfserr_perm;
	if (!flen)
		goto out;
	err = nfserr_exist;
	if (isdotent(fname, flen))
		goto out;

	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE);
	if (err)
		goto out;

	dentry = fhp->fh_dentry;
	dirp = dentry->d_inode;

	err = nfserr_notdir;
	if(!dirp->i_op || !dirp->i_op->lookup)
		goto out;
	/*
	 * Check whether the response file handle has been verified yet.
	 * If it has, the parent directory should already be locked.
	 */
	if (!resfhp->fh_dentry) {
		/* called from nfsd_proc_mkdir, or possibly nfsd3_proc_create */
		fh_lock(fhp);
		dchild = lookup_one_len(fname, dentry, flen);
		err = PTR_ERR(dchild);
		if (IS_ERR(dchild))
			goto out_nfserr;
		err = fh_compose(resfhp, fhp->fh_export, dchild, fhp);
		if (err)
			goto out;
	} else {
		/* called from nfsd_proc_create */
		dchild = dget(resfhp->fh_dentry);
		if (!fhp->fh_locked) {
			/* not actually possible */
			printk(KERN_ERR
				"nfsd_create: parent %s/%s not locked!\n",
				dentry->d_parent->d_name.name,
				dentry->d_name.name);
			err = -EIO;
			goto out;
		}
	}
	/*
	 * Make sure the child dentry is still negative ...
	 */
	err = nfserr_exist;
	if (dchild->d_inode) {
		dprintk("nfsd_create: dentry %s/%s not negative!\n",
			dentry->d_name.name, dchild->d_name.name);
		goto out; 
	}

	if (!(iap->ia_valid & ATTR_MODE))
		iap->ia_mode = 0;
	iap->ia_mode = (iap->ia_mode & S_IALLUGO) | type;

	/*
	 * Get the dir op function pointer.
	 */
	err = nfserr_perm;
	switch (type) {
	case S_IFREG:
		err = vfs_create(dirp, dchild, iap->ia_mode, NULL);
		break;
	case S_IFDIR:
		err = vfs_mkdir(dirp, dchild, iap->ia_mode);
		break;
	case S_IFCHR:
	case S_IFBLK:
	case S_IFIFO:
	case S_IFSOCK:
		err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev);
		break;
	default:
	        printk("nfsd: bad file type %o in nfsd_create\n", type);
		err = -EINVAL;
	}
	if (err < 0)
		goto out_nfserr;

	if (EX_ISSYNC(fhp->fh_export)) {
		nfsd_sync_dir(dentry);
		write_inode_now(dchild->d_inode, 1);
	}


	/* 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.
	 */
	err = 0;
	if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0)
		err = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0);
	/*
	 * Update the file handle to get the new inode info.
	 */
	if (!err)
		err = fh_update(resfhp);
out:
	if (dchild && !IS_ERR(dchild))
		dput(dchild);
	return err;

out_nfserr:
	err = nfserrno(err);
	goto out;
}

#ifdef CONFIG_NFSD_V3
/*
 * NFSv3 version of nfsd_create
 */
int
nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
		char *fname, int flen, struct iattr *iap,
		struct svc_fh *resfhp, int createmode, u32 *verifier,
	        int *truncp)
{
	struct dentry	*dentry, *dchild = NULL;
	struct inode	*dirp;
	int		err;
	__u32		v_mtime=0, v_atime=0;
	int		v_mode=0;

	err = nfserr_perm;
	if (!flen)
		goto out;
	err = nfserr_exist;
	if (isdotent(fname, flen))
		goto out;
	if (!(iap->ia_valid & ATTR_MODE))
		iap->ia_mode = 0;
	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE);
	if (err)
		goto out;

	dentry = fhp->fh_dentry;
	dirp = dentry->d_inode;

	/* Get all the sanity checks out of the way before
	 * we lock the parent. */
	err = nfserr_notdir;
	if(!dirp->i_op || !dirp->i_op->lookup)
		goto out;
	fh_lock(fhp);

	/*
	 * Compose the response file handle.
	 */
	dchild = lookup_one_len(fname, dentry, flen);
	err = PTR_ERR(dchild);
	if (IS_ERR(dchild))
		goto out_nfserr;

	err = fh_compose(resfhp, fhp->fh_export, dchild, fhp);
	if (err)
		goto out;

	if (createmode == NFS3_CREATE_EXCLUSIVE) {
		/* while the verifier would fit in mtime+atime,
		 * solaris7 gets confused (bugid 4218508) if these have
		 * the high bit set, so we use the mode as well