freebsd-dev/sys/ufs/lfs/lfs_vnops.c
David Greenman f6b04d2bfb Changes from John Dyson and myself:
Fixed remaining known bugs in the buffer IO and VM system.

vfs_bio.c:
Fixed some race conditions and locking bugs. Improved performance
by removing some (now) unnecessary code and fixing some broken
logic.
Fixed process accounting of # of FS outputs.
Properly handle NFS interrupts (B_EINTR).

(various)
Replaced calls to clrbuf() with calls to an optimized routine
called vfs_bio_clrbuf().

(various FS sync)
Sync out modified vnode_pager backed pages.

ffs_vnops.c:
Do two passes: Sync out file data first, then indirect blocks.

vm_fault.c:
Fixed deadly embrace caused by acquiring locks in the wrong order.

vnode_pager.c:
Changed to use buffer I/O system for writing out modified pages. This
should fix the problem with the modification date previous not getting
updated. Also dramatically simplifies the code. Note that this is
going to change in the future and be implemented via VOP_PUTPAGES().

vm_object.c:
Fixed a pile of bugs related to cleaning (vnode) objects. The performance
of vm_object_page_clean() is terrible when dealing with huge objects,
but this will change when we implement a binary tree to keep the object
pages sorted.

vm_pageout.c:
Fixed broken clustering of pageouts. Fixed race conditions and other
lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00

505 lines
15 KiB
C

/*
* Copyright (c) 1986, 1989, 1991, 1993
* 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.
*
* @(#)lfs_vnops.c 8.5 (Berkeley) 12/30/93
* $Id: lfs_vnops.c,v 1.7 1995/03/28 07:58:06 bde Exp $
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/namei.h>
#include <sys/resourcevar.h>
#include <sys/signalvar.h>
#include <sys/kernel.h>
#include <sys/file.h>
#include <sys/stat.h>
#include <sys/buf.h>
#include <sys/proc.h>
#include <sys/conf.h>
#include <sys/mount.h>
#include <sys/vnode.h>
#include <sys/malloc.h>
#include <vm/vm.h>
#include <vm/vm_page.h>
#include <vm/vm_object.h>
#include <miscfs/specfs/specdev.h>
#include <miscfs/fifofs/fifo.h>
#include <ufs/ufs/quota.h>
#include <ufs/ufs/inode.h>
#include <ufs/ufs/dir.h>
#include <ufs/ufs/ufsmount.h>
#include <ufs/ufs/ufs_extern.h>
#include <ufs/lfs/lfs.h>
#include <ufs/lfs/lfs_extern.h>
/* Global vfs data structures for lfs. */
int (**lfs_vnodeop_p)();
struct vnodeopv_entry_desc lfs_vnodeop_entries[] = {
{ &vop_default_desc, vn_default_error },
{ &vop_lookup_desc, ufs_lookup }, /* lookup */
{ &vop_create_desc, ufs_create }, /* create */
{ &vop_mknod_desc, ufs_mknod }, /* mknod */
{ &vop_open_desc, ufs_open }, /* open */
{ &vop_close_desc, lfs_close }, /* close */
{ &vop_access_desc, ufs_access }, /* access */
{ &vop_getattr_desc, lfs_getattr }, /* getattr */
{ &vop_setattr_desc, ufs_setattr }, /* setattr */
{ &vop_read_desc, lfs_read }, /* read */
{ &vop_write_desc, lfs_write }, /* write */
{ &vop_ioctl_desc, ufs_ioctl }, /* ioctl */
{ &vop_select_desc, ufs_select }, /* select */
{ &vop_mmap_desc, ufs_mmap }, /* mmap */
{ &vop_fsync_desc, lfs_fsync }, /* fsync */
{ &vop_seek_desc, ufs_seek }, /* seek */
{ &vop_remove_desc, ufs_remove }, /* remove */
{ &vop_link_desc, ufs_link }, /* link */
{ &vop_rename_desc, ufs_rename }, /* rename */
{ &vop_mkdir_desc, ufs_mkdir }, /* mkdir */
{ &vop_rmdir_desc, ufs_rmdir }, /* rmdir */
{ &vop_symlink_desc, ufs_symlink }, /* symlink */
{ &vop_readdir_desc, ufs_readdir }, /* readdir */
{ &vop_readlink_desc, ufs_readlink }, /* readlink */
{ &vop_abortop_desc, ufs_abortop }, /* abortop */
{ &vop_inactive_desc, lfs_inactive }, /* inactive */
{ &vop_reclaim_desc, ufs_reclaim }, /* reclaim */
{ &vop_lock_desc, ufs_lock }, /* lock */
{ &vop_unlock_desc, ufs_unlock }, /* unlock */
{ &vop_bmap_desc, ufs_bmap }, /* bmap */
{ &vop_strategy_desc, ufs_strategy }, /* strategy */
{ &vop_print_desc, ufs_print }, /* print */
{ &vop_islocked_desc, ufs_islocked }, /* islocked */
{ &vop_pathconf_desc, ufs_pathconf }, /* pathconf */
{ &vop_advlock_desc, ufs_advlock }, /* advlock */
{ &vop_blkatoff_desc, lfs_blkatoff }, /* blkatoff */
{ &vop_valloc_desc, lfs_valloc }, /* valloc */
{ &vop_vfree_desc, lfs_vfree }, /* vfree */
{ &vop_truncate_desc, lfs_truncate }, /* truncate */
{ &vop_update_desc, lfs_update }, /* update */
{ &vop_bwrite_desc, lfs_bwrite }, /* bwrite */
{ (struct vnodeop_desc*)NULL, (int(*)())NULL }
};
struct vnodeopv_desc lfs_vnodeop_opv_desc =
{ &lfs_vnodeop_p, lfs_vnodeop_entries };
int (**lfs_specop_p)();
struct vnodeopv_entry_desc lfs_specop_entries[] = {
{ &vop_default_desc, vn_default_error },
{ &vop_lookup_desc, spec_lookup }, /* lookup */
{ &vop_create_desc, spec_create }, /* create */
{ &vop_mknod_desc, spec_mknod }, /* mknod */
{ &vop_open_desc, spec_open }, /* open */
{ &vop_close_desc, ufsspec_close }, /* close */
{ &vop_access_desc, ufs_access }, /* access */
{ &vop_getattr_desc, lfs_getattr }, /* getattr */
{ &vop_setattr_desc, ufs_setattr }, /* setattr */
{ &vop_read_desc, ufsspec_read }, /* read */
{ &vop_write_desc, ufsspec_write }, /* write */
{ &vop_ioctl_desc, spec_ioctl }, /* ioctl */
{ &vop_select_desc, spec_select }, /* select */
{ &vop_mmap_desc, spec_mmap }, /* mmap */
{ &vop_fsync_desc, spec_fsync }, /* fsync */
{ &vop_seek_desc, spec_seek }, /* seek */
{ &vop_remove_desc, spec_remove }, /* remove */
{ &vop_link_desc, spec_link }, /* link */
{ &vop_rename_desc, spec_rename }, /* rename */
{ &vop_mkdir_desc, spec_mkdir }, /* mkdir */
{ &vop_rmdir_desc, spec_rmdir }, /* rmdir */
{ &vop_symlink_desc, spec_symlink }, /* symlink */
{ &vop_readdir_desc, spec_readdir }, /* readdir */
{ &vop_readlink_desc, spec_readlink }, /* readlink */
{ &vop_abortop_desc, spec_abortop }, /* abortop */
{ &vop_inactive_desc, lfs_inactive }, /* inactive */
{ &vop_reclaim_desc, ufs_reclaim }, /* reclaim */
{ &vop_lock_desc, ufs_lock }, /* lock */
{ &vop_unlock_desc, ufs_unlock }, /* unlock */
{ &vop_bmap_desc, spec_bmap }, /* bmap */
{ &vop_strategy_desc, spec_strategy }, /* strategy */
{ &vop_print_desc, ufs_print }, /* print */
{ &vop_islocked_desc, ufs_islocked }, /* islocked */
{ &vop_pathconf_desc, spec_pathconf }, /* pathconf */
{ &vop_advlock_desc, spec_advlock }, /* advlock */
{ &vop_blkatoff_desc, spec_blkatoff }, /* blkatoff */
{ &vop_valloc_desc, spec_valloc }, /* valloc */
{ &vop_vfree_desc, lfs_vfree }, /* vfree */
{ &vop_truncate_desc, spec_truncate }, /* truncate */
{ &vop_update_desc, lfs_update }, /* update */
{ &vop_bwrite_desc, lfs_bwrite }, /* bwrite */
{ (struct vnodeop_desc*)NULL, (int(*)())NULL }
};
struct vnodeopv_desc lfs_specop_opv_desc =
{ &lfs_specop_p, lfs_specop_entries };
int (**lfs_fifoop_p)();
struct vnodeopv_entry_desc lfs_fifoop_entries[] = {
{ &vop_default_desc, vn_default_error },
{ &vop_lookup_desc, fifo_lookup }, /* lookup */
{ &vop_create_desc, fifo_create }, /* create */
{ &vop_mknod_desc, fifo_mknod }, /* mknod */
{ &vop_open_desc, fifo_open }, /* open */
{ &vop_close_desc, ufsfifo_close }, /* close */
{ &vop_access_desc, ufs_access }, /* access */
{ &vop_getattr_desc, lfs_getattr }, /* getattr */
{ &vop_setattr_desc, ufs_setattr }, /* setattr */
{ &vop_read_desc, ufsfifo_read }, /* read */
{ &vop_write_desc, ufsfifo_write }, /* write */
{ &vop_ioctl_desc, fifo_ioctl }, /* ioctl */
{ &vop_select_desc, fifo_select }, /* select */
{ &vop_mmap_desc, fifo_mmap }, /* mmap */
{ &vop_fsync_desc, fifo_fsync }, /* fsync */
{ &vop_seek_desc, fifo_seek }, /* seek */
{ &vop_remove_desc, fifo_remove }, /* remove */
{ &vop_link_desc, fifo_link }, /* link */
{ &vop_rename_desc, fifo_rename }, /* rename */
{ &vop_mkdir_desc, fifo_mkdir }, /* mkdir */
{ &vop_rmdir_desc, fifo_rmdir }, /* rmdir */
{ &vop_symlink_desc, fifo_symlink }, /* symlink */
{ &vop_readdir_desc, fifo_readdir }, /* readdir */
{ &vop_readlink_desc, fifo_readlink }, /* readlink */
{ &vop_abortop_desc, fifo_abortop }, /* abortop */
{ &vop_inactive_desc, lfs_inactive }, /* inactive */
{ &vop_reclaim_desc, ufs_reclaim }, /* reclaim */
{ &vop_lock_desc, ufs_lock }, /* lock */
{ &vop_unlock_desc, ufs_unlock }, /* unlock */
{ &vop_bmap_desc, fifo_bmap }, /* bmap */
{ &vop_strategy_desc, fifo_strategy }, /* strategy */
{ &vop_print_desc, ufs_print }, /* print */
{ &vop_islocked_desc, ufs_islocked }, /* islocked */
{ &vop_pathconf_desc, fifo_pathconf }, /* pathconf */
{ &vop_advlock_desc, fifo_advlock }, /* advlock */
{ &vop_blkatoff_desc, fifo_blkatoff }, /* blkatoff */
{ &vop_valloc_desc, fifo_valloc }, /* valloc */
{ &vop_vfree_desc, lfs_vfree }, /* vfree */
{ &vop_truncate_desc, fifo_truncate }, /* truncate */
{ &vop_update_desc, lfs_update }, /* update */
{ &vop_bwrite_desc, lfs_bwrite }, /* bwrite */
{ (struct vnodeop_desc*)NULL, (int(*)())NULL }
};
struct vnodeopv_desc lfs_fifoop_opv_desc =
{ &lfs_fifoop_p, lfs_fifoop_entries };
VNODEOP_SET(lfs_vnodeop_opv_desc);
VNODEOP_SET(lfs_specop_opv_desc);
VNODEOP_SET(lfs_fifoop_opv_desc);
#define LFS_READWRITE
#include <ufs/ufs/ufs_readwrite.c>
#undef LFS_READWRITE
/*
* Synch an open file.
*/
/* ARGSUSED */
int
lfs_fsync(ap)
struct vop_fsync_args /* {
struct vnode *a_vp;
struct ucred *a_cred;
int a_waitfor;
struct proc *a_p;
} */ *ap;
{
struct timeval tv;
int error;
tv = time;
/*
* If the vnode has an object, then flush all of the dirty pages
* into the buffer cache.
*/
if (ap->a_vp->v_vmdata)
_vm_object_page_clean((vm_object_t)ap->a_vp->v_vmdata, 0, 0, 0);
error = (VOP_UPDATE(ap->a_vp, &tv, &tv,
ap->a_waitfor == MNT_WAIT ? LFS_SYNC : 0));
if(ap->a_waitfor == MNT_WAIT && ap->a_vp->v_dirtyblkhd.lh_first != NULL)
panic("lfs_fsync: dirty bufs");
return( error );
}
/*
* These macros are used to bracket UFS directory ops, so that we can
* identify all the pages touched during directory ops which need to
* be ordered and flushed atomically, so that they may be recovered.
*/
#define SET_DIROP(fs) { \
if ((fs)->lfs_writer) \
tsleep(&(fs)->lfs_dirops, PRIBIO + 1, "lfs_dirop", 0); \
++(fs)->lfs_dirops; \
(fs)->lfs_doifile = 1; \
}
#define SET_ENDOP(fs) { \
--(fs)->lfs_dirops; \
if (!(fs)->lfs_dirops) \
wakeup(&(fs)->lfs_writer); \
}
#define MARK_VNODE(dvp) (dvp)->v_flag |= VDIROP
int
lfs_symlink(ap)
struct vop_symlink_args /* {
struct vnode *a_dvp;
struct vnode **a_vpp;
struct componentname *a_cnp;
struct vattr *a_vap;
char *a_target;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_dvp)->i_lfs);
MARK_VNODE(ap->a_dvp);
ret = ufs_symlink(ap);
SET_ENDOP(VTOI(ap->a_dvp)->i_lfs);
return (ret);
}
int
lfs_mknod(ap)
struct vop_mknod_args /* {
struct vnode *a_dvp;
struct vnode **a_vpp;
struct componentname *a_cnp;
struct vattr *a_vap;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_dvp)->i_lfs);
MARK_VNODE(ap->a_dvp);
ret = ufs_mknod(ap);
SET_ENDOP(VTOI(ap->a_dvp)->i_lfs);
return (ret);
}
int
lfs_create(ap)
struct vop_create_args /* {
struct vnode *a_dvp;
struct vnode **a_vpp;
struct componentname *a_cnp;
struct vattr *a_vap;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_dvp)->i_lfs);
MARK_VNODE(ap->a_dvp);
ret = ufs_create(ap);
SET_ENDOP(VTOI(ap->a_dvp)->i_lfs);
return (ret);
}
int
lfs_mkdir(ap)
struct vop_mkdir_args /* {
struct vnode *a_dvp;
struct vnode **a_vpp;
struct componentname *a_cnp;
struct vattr *a_vap;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_dvp)->i_lfs);
MARK_VNODE(ap->a_dvp);
ret = ufs_mkdir(ap);
SET_ENDOP(VTOI(ap->a_dvp)->i_lfs);
return (ret);
}
int
lfs_remove(ap)
struct vop_remove_args /* {
struct vnode *a_dvp;
struct vnode *a_vp;
struct componentname *a_cnp;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_dvp)->i_lfs);
MARK_VNODE(ap->a_dvp);
MARK_VNODE(ap->a_vp);
ret = ufs_remove(ap);
SET_ENDOP(VTOI(ap->a_dvp)->i_lfs);
return (ret);
}
int
lfs_rmdir(ap)
struct vop_rmdir_args /* {
struct vnodeop_desc *a_desc;
struct vnode *a_dvp;
struct vnode *a_vp;
struct componentname *a_cnp;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_dvp)->i_lfs);
MARK_VNODE(ap->a_dvp);
MARK_VNODE(ap->a_vp);
ret = ufs_rmdir(ap);
SET_ENDOP(VTOI(ap->a_dvp)->i_lfs);
return (ret);
}
int
lfs_link(ap)
struct vop_link_args /* {
struct vnode *a_vp;
struct vnode *a_tdvp;
struct componentname *a_cnp;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_vp)->i_lfs);
MARK_VNODE(ap->a_vp);
ret = ufs_link(ap);
SET_ENDOP(VTOI(ap->a_vp)->i_lfs);
return (ret);
}
int
lfs_rename(ap)
struct vop_rename_args /* {
struct vnode *a_fdvp;
struct vnode *a_fvp;
struct componentname *a_fcnp;
struct vnode *a_tdvp;
struct vnode *a_tvp;
struct componentname *a_tcnp;
} */ *ap;
{
int ret;
SET_DIROP(VTOI(ap->a_fdvp)->i_lfs);
MARK_VNODE(ap->a_fdvp);
MARK_VNODE(ap->a_tdvp);
ret = ufs_rename(ap);
SET_ENDOP(VTOI(ap->a_fdvp)->i_lfs);
return (ret);
}
/* XXX hack to avoid calling ITIMES in getattr */
int
lfs_getattr(ap)
struct vop_getattr_args /* {
struct vnode *a_vp;
struct vattr *a_vap;
struct ucred *a_cred;
struct proc *a_p;
} */ *ap;
{
register struct vnode *vp = ap->a_vp;
register struct inode *ip = VTOI(vp);
register struct vattr *vap = ap->a_vap;
/*
* Copy from inode table
*/
vap->va_fsid = ip->i_dev;
vap->va_fileid = ip->i_number;
vap->va_mode = ip->i_mode & ~IFMT;
vap->va_nlink = ip->i_nlink;
vap->va_uid = ip->i_uid;
vap->va_gid = ip->i_gid;
vap->va_rdev = (dev_t)ip->i_rdev;
vap->va_size = ip->i_din.di_size;
vap->va_atime = ip->i_atime;
vap->va_mtime = ip->i_mtime;
vap->va_ctime = ip->i_ctime;
vap->va_flags = ip->i_flags;
vap->va_gen = ip->i_gen;
/* this doesn't belong here */
if (vp->v_type == VBLK)
vap->va_blocksize = BLKDEV_IOSIZE;
else if (vp->v_type == VCHR)
vap->va_blocksize = MAXBSIZE;
else
vap->va_blocksize = vp->v_mount->mnt_stat.f_iosize;
vap->va_bytes = dbtob(ip->i_blocks);
vap->va_type = vp->v_type;
vap->va_filerev = ip->i_modrev;
return (0);
}
/*
* Close called
*
* XXX -- we were using ufs_close, but since it updates the
* times on the inode, we might need to bump the uinodes
* count.
*/
/* ARGSUSED */
int
lfs_close(ap)
struct vop_close_args /* {
struct vnode *a_vp;
int a_fflag;
struct ucred *a_cred;
struct proc *a_p;
} */ *ap;
{
register struct vnode *vp = ap->a_vp;
register struct inode *ip = VTOI(vp);
int mod;
if (vp->v_usecount > 1 && !(ip->i_flag & IN_LOCKED)) {
mod = ip->i_flag & IN_MODIFIED;
ITIMES(ip, &time, &time);
if (!mod && ip->i_flag & IN_MODIFIED)
ip->i_lfs->lfs_uinodes++;
}
return (0);
}
/*
* Stub inactive routine that avoids calling ufs_inactive in some cases.
*/
int
lfs_inactive(ap)
struct vop_inactive_args /* {
struct vnode *a_vp;
} */ *ap;
{
if (ap->a_vp->v_flag & VNINACT) {
return(0);
}
return (ufs_inactive(ap));
}