ffs_vfsops.c

早期freebsd实现

/*
 * Copyright (c) 1989, 1991, 1993, 1994
 *	The Regents of the University of California.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)ffs_vfsops.c	8.8 (Berkeley) 4/18/94
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/namei.h>
#include <sys/proc.h>
#include <sys/kernel.h>
#include <sys/vnode.h>
#include <sys/socket.h>
#include <sys/mount.h>
#include <sys/buf.h>
#include <sys/mbuf.h>
#include <sys/file.h>
#include <sys/disklabel.h>
#include <sys/ioctl.h>
#include <sys/errno.h>
#include <sys/malloc.h>

#include <miscfs/specfs/specdev.h>

#include <ufs/ufs/quota.h>
#include <ufs/ufs/ufsmount.h>
#include <ufs/ufs/inode.h>
#include <ufs/ufs/ufs_extern.h>

#include <ufs/ffs/fs.h>
#include <ufs/ffs/ffs_extern.h>

int ffs_sbupdate __P((struct ufsmount *, int));

struct vfsops ufs_vfsops = {
	ffs_mount,
	ufs_start,
	ffs_unmount,
	ufs_root,
	ufs_quotactl,
	ffs_statfs,
	ffs_sync,
	ffs_vget,
	ffs_fhtovp,
	ffs_vptofh,
	ffs_init,
};

extern u_long nextgennumber;

/*
 * Called by main() when ufs is going to be mounted as root.
 *
 * Name is updated by mount(8) after booting.
 */
#define ROOTNAME	"root_device"

ffs_mountroot()
{
	extern struct vnode *rootvp;
	register struct fs *fs;
	register struct mount *mp;
	struct proc *p = curproc;	/* XXX */
	struct ufsmount *ump;
	u_int size;
	int error;
	
	/*
	 * Get vnodes for swapdev and rootdev.
	 */
	if (bdevvp(swapdev, &swapdev_vp) || bdevvp(rootdev, &rootvp))
		panic("ffs_mountroot: can't setup bdevvp's");

	mp = malloc((u_long)sizeof(struct mount), M_MOUNT, M_WAITOK);
	bzero((char *)mp, (u_long)sizeof(struct mount));
	mp->mnt_op = &ufs_vfsops;
	mp->mnt_flag = MNT_RDONLY;
	if (error = ffs_mountfs(rootvp, mp, p)) {
		free(mp, M_MOUNT);
		return (error);
	}
	if (error = vfs_lock(mp)) {
		(void)ffs_unmount(mp, 0, p);
		free(mp, M_MOUNT);
		return (error);
	}
	TAILQ_INSERT_TAIL(&mountlist, mp, mnt_list);
	mp->mnt_flag |= MNT_ROOTFS;
	mp->mnt_vnodecovered = NULLVP;
	ump = VFSTOUFS(mp);
	fs = ump->um_fs;
	bzero(fs->fs_fsmnt, sizeof(fs->fs_fsmnt));
	fs->fs_fsmnt[0] = '/';
	bcopy((caddr_t)fs->fs_fsmnt, (caddr_t)mp->mnt_stat.f_mntonname,
	    MNAMELEN);
	(void) copystr(ROOTNAME, mp->mnt_stat.f_mntfromname, MNAMELEN - 1,
	    &size);
	bzero(mp->mnt_stat.f_mntfromname + size, MNAMELEN - size);
	(void)ffs_statfs(mp, &mp->mnt_stat, p);
	vfs_unlock(mp);
	inittodr(fs->fs_time);
	return (0);
}

/*
 * VFS Operations.
 *
 * mount system call
 */
int
ffs_mount(mp, path, data, ndp, p)
	register struct mount *mp;
	char *path;
	caddr_t data;
	struct nameidata *ndp;
	struct proc *p;
{
	struct vnode *devvp;
	struct ufs_args args;
	struct ufsmount *ump;
	register struct fs *fs;
	u_int size;
	int error, flags;

	if (error = copyin(data, (caddr_t)&args, sizeof (struct ufs_args)))
		return (error);
	/*
	 * If updating, check whether changing from read-only to
	 * read/write; if there is no device name, that's all we do.
	 */
	if (mp->mnt_flag & MNT_UPDATE) {
		ump = VFSTOUFS(mp);
		fs = ump->um_fs;
		error = 0;
		if (fs->fs_ronly == 0 && (mp->mnt_flag & MNT_RDONLY)) {
			flags = WRITECLOSE;
			if (mp->mnt_flag & MNT_FORCE)
				flags |= FORCECLOSE;
			if (vfs_busy(mp))
				return (EBUSY);
			error = ffs_flushfiles(mp, flags, p);
			vfs_unbusy(mp);
		}
		if (!error && (mp->mnt_flag & MNT_RELOAD))
			error = ffs_reload(mp, ndp->ni_cnd.cn_cred, p);
		if (error)
			return (error);
		if (fs->fs_ronly && (mp->mnt_flag & MNT_WANTRDWR))
			fs->fs_ronly = 0;
		if (args.fspec == 0) {
			/*
			 * Process export requests.
			 */
			return (vfs_export(mp, &ump->um_export, &args.export));
		}
	}
	/*
	 * Not an update, or updating the name: look up the name
	 * and verify that it refers to a sensible block device.
	 */
	NDINIT(ndp, LOOKUP, FOLLOW, UIO_USERSPACE, args.fspec, p);
	if (error = namei(ndp))
		return (error);
	devvp = ndp->ni_vp;

	if (devvp->v_type != VBLK) {
		vrele(devvp);
		return (ENOTBLK);
	}
	if (major(devvp->v_rdev) >= nblkdev) {
		vrele(devvp);
		return (ENXIO);
	}
	if ((mp->mnt_flag & MNT_UPDATE) == 0)
		error = ffs_mountfs(devvp, mp, p);
	else {
		if (devvp != ump->um_devvp)
			error = EINVAL;	/* needs translation */
		else
			vrele(devvp);
	}
	if (error) {
		vrele(devvp);
		return (error);
	}
	ump = VFSTOUFS(mp);
	fs = ump->um_fs;
	(void) copyinstr(path, fs->fs_fsmnt, sizeof(fs->fs_fsmnt) - 1, &size);
	bzero(fs->fs_fsmnt + size, sizeof(fs->fs_fsmnt) - size);
	bcopy((caddr_t)fs->fs_fsmnt, (caddr_t)mp->mnt_stat.f_mntonname,
	    MNAMELEN);
	(void) copyinstr(args.fspec, mp->mnt_stat.f_mntfromname, MNAMELEN - 1, 
	    &size);
	bzero(mp->mnt_stat.f_mntfromname + size, MNAMELEN - size);
	(void)ffs_statfs(mp, &mp->mnt_stat, p);
	return (0);
}

/*
 * Reload all incore data for a filesystem (used after running fsck on
 * the root filesystem and finding things to fix). The filesystem must
 * be mounted read-only.
 *
 * Things to do to update the mount:
 *	1) invalidate all cached meta-data.
 *	2) re-read superblock from disk.
 *	3) re-read summary information from disk.
 *	4) invalidate all inactive vnodes.
 *	5) invalidate all cached file data.
 *	6) re-read inode data for all active vnodes.
 */
ffs_reload(mountp, cred, p)
	register struct mount *mountp;
	struct ucred *cred;
	struct proc *p;
{
	register struct vnode *vp, *nvp, *devvp;
	struct inode *ip;
	struct csum *space;
	struct buf *bp;
	struct fs *fs;
	int i, blks, size, error;

	if ((mountp->mnt_flag & MNT_RDONLY) == 0)
		return (EINVAL);
	/*
	 * Step 1: invalidate all cached meta-data.
	 */
	devvp = VFSTOUFS(mountp)->um_devvp;
	if (vinvalbuf(devvp, 0, cred, p, 0, 0))
		panic("ffs_reload: dirty1");
	/*
	 * Step 2: re-read superblock from disk.
	 */
	if (error = bread(devvp, SBLOCK, SBSIZE, NOCRED, &bp))
		return (error);
	fs = (struct fs *)bp->b_data;
	if (fs->fs_magic != FS_MAGIC || fs->fs_bsize > MAXBSIZE ||
	    fs->fs_bsize < sizeof(struct fs)) {
		brelse(bp);
		return (EIO);		/* XXX needs translation */
	}
	fs = VFSTOUFS(mountp)->um_fs;
	bcopy(&fs->fs_csp[0], &((struct fs *)bp->b_data)->fs_csp[0],
	    sizeof(fs->fs_csp));
	bcopy(bp->b_data, fs, (u_int)fs->fs_sbsize);
	if (fs->fs_sbsize < SBSIZE)
		bp->b_flags |= B_INVAL;
	brelse(bp);
	ffs_oldfscompat(fs);
	/*
	 * Step 3: re-read summary information from disk.
	 */
	blks = howmany(fs->fs_cssize, fs->fs_fsize);
	space = fs->fs_csp[0];
	for (i = 0; i < blks; i += fs->fs_frag) {
		size = fs->fs_bsize;
		if (i + fs->fs_frag > blks)
			size = (blks - i) * fs->fs_fsize;
		if (error = bread(devvp, fsbtodb(fs, fs->fs_csaddr + i), size,
		    NOCRED, &bp))
			return (error);
		bcopy(bp->b_data, fs->fs_csp[fragstoblks(fs, i)], (u_int)size);
		brelse(bp);
	}
loop:
	for (vp = mountp->mnt_vnodelist.lh_first; vp != NULL; vp = nvp) {
		nvp = vp->v_mntvnodes.le_next;
		/*
		 * Step 4: invalidate all inactive vnodes.
		 */
		if (vp->v_usecount == 0) {
			vgone(vp);
			continue;
		}
		/*
		 * Step 5: invalidate all cached file data.
		 */
		if (vget(vp, 1))
			goto loop;
		if (vinvalbuf(vp, 0, cred, p, 0, 0))
			panic("ffs_reload: dirty2");
		/*
		 * Step 6: re-read inode data for all active vnodes.
		 */
		ip = VTOI(vp);
		if (error =
		    bread(devvp, fsbtodb(fs, ino_to_fsba(fs, ip->i_number)),
		    (int)fs->fs_bsize, NOCRED, &bp)) {
			vput(vp);
			return (error);
		}
		ip->i_din = *((struct dinode *)bp->b_data +
		    ino_to_fsbo(fs, ip->i_number));
		brelse(bp);
		vput(vp);
		if (vp->v_mount != mountp)
			goto loop;
	}
	return (0);
}

/*
 * Common code for mount and mountroot
 */
int
ffs_mountfs(devvp, mp, p)
	register struct vnode *devvp;
	struct mount *mp;
	struct proc *p;
{
	register struct ufsmount *ump;
	struct buf *bp;
	register struct fs *fs;
	dev_t dev = devvp->v_rdev;
	struct partinfo dpart;
	caddr_t base, space;
	int havepart = 0, blks;
	int error, i, size;
	int ronly;
	extern struct vnode *rootvp;

	/*
	 * Disallow multiple mounts of the same device.
	 * Disallow mounting of a device that is currently in use
	 * (except for root, which might share swap device for miniroot).
	 * Flush out any old buffers remaining from a previous use.
	 */
	if (error = vfs_mountedon(devvp))
		return (error);
	if (vcount(devvp) > 1 && devvp != rootvp)
		return (EBUSY);
	if (error = vinvalbuf(devvp, V_SAVE, p->p_ucred, p, 0, 0))
		return (error);

	ronly = (mp->mnt_flag & MNT_RDONLY) != 0;
	if (error = VOP_OPEN(devvp, ronly ? FREAD : FREAD|FWRITE, FSCRED, p))
		return (error);
	if (VOP_IOCTL(devvp, DIOCGPART, (caddr_t)&dpart, FREAD, NOCRED, p) != 0)
		size = DEV_BSIZE;
	else {
		havepart = 1;
		size = dpart.disklab->d_secsize;
	}

	bp = NULL;
	ump = NULL;
	if (error = bread(devvp, SBLOCK, SBSIZE, NOCRED, &bp))
		goto out;
	fs = (struct fs *)bp->b_data;
	if (fs->fs_magic != FS_MAGIC || fs->fs_bsize > MAXBSIZE ||
	    fs->fs_bsize < sizeof(struct fs)) {
		error = EINVAL;		/* XXX needs translation */
		goto out;
	}
	ump = malloc(sizeof *ump, M_UFSMNT, M_WAITOK);
	bzero((caddr_t)ump, sizeof *ump);
	ump->um_fs = malloc((u_long)fs->fs_sbsize, M_UFSMNT,
	    M_WAITOK);
	bcopy(bp->b_data, ump->um_fs, (u_int)fs->fs_sbsize);
	if (fs->fs_sbsize < SBSIZE)
		bp->b_flags |= B_INVAL;
	brelse(bp);
	bp = NULL;
	fs = ump->um_fs;
	fs->fs_ronly = ronly;
	if (ronly == 0)
		fs->fs_fmod = 1;
	blks = howmany(fs->fs_cssize, fs->fs_fsize);
	base = space = malloc((u_long)fs->fs_cssize, M_UFSMNT,
	    M_WAITOK);
	for (i = 0; i < blks; i += fs->fs_frag) {
		size = fs->fs_bsize;
		if (i + fs->fs_frag > blks)
			size = (blks - i) * fs->fs_fsize;
		error = bread(devvp, fsbtodb(fs, fs->fs_csaddr + i), size,
			NOCRED, &bp);
		if (error) {
			free(base, M_UFSMNT);
			goto out;
		}
		bcopy(bp->b_data, space, (u_int)size);
		fs->fs_csp[fragstoblks(fs, i)] = (struct csum *)space;
		space += size;
		brelse(bp);
		bp = NULL;
	}
	mp->mnt_data = (qaddr_t)ump;
	mp->mnt_stat.f_fsid.val[0] = (long)dev;
	mp->mnt_stat.f_fsid.val[1] = MOUNT_UFS;
	mp->mnt_maxsymlinklen = fs->fs_maxsymlinklen;
	mp->mnt_flag |= MNT_LOCAL;
	ump->um_mountp = mp;