freebsd-skq/sys/kern/vfs_default.c
John Dyson 8f9110f6a1 This mega-commit is meant to fix numerous interrelated problems. There
has been some bitrot and incorrect assumptions in the vfs_bio code.  These
problems have manifest themselves worse on NFS type filesystems, but can
still affect local filesystems under certain circumstances.  Most of
the problems have involved mmap consistancy, and as a side-effect broke
the vfs.ioopt code.  This code might have been committed seperately, but
almost everything is interrelated.

1)	Allow (pmap_object_init_pt) prefaulting of buffer-busy pages that
	are fully valid.
2)	Rather than deactivating erroneously read initial (header) pages in
	kern_exec, we now free them.
3)	Fix the rundown of non-VMIO buffers that are in an inconsistent
	(missing vp) state.
4)	Fix the disassociation of pages from buffers in brelse.  The previous
	code had rotted and was faulty in a couple of important circumstances.
5)	Remove a gratuitious buffer wakeup in vfs_vmio_release.
6)	Remove a crufty and currently unused cluster mechanism for VBLK
	files in vfs_bio_awrite.  When the code is functional, I'll add back
	a cleaner version.
7)	The page busy count wakeups assocated with the buffer cache usage were
	incorrectly cleaned up in a previous commit by me.  Revert to the
	original, correct version, but with a cleaner implementation.
8)	The cluster read code now tries to keep data associated with buffers
	more aggressively (without breaking the heuristics) when it is presumed
	that the read data (buffers) will be soon needed.
9)	Change to filesystem lockmgr locks so that they use LK_NOPAUSE.  The
	delay loop waiting is not useful for filesystem locks, due to the
	length of the time intervals.
10)	Correct and clean-up spec_getpages.
11)	Implement a fully functional nfs_getpages, nfs_putpages.
12)	Fix nfs_write so that modifications are coherent with the NFS data on
	the server disk (at least as well as NFS seems to allow.)
13)	Properly support MS_INVALIDATE on NFS.
14)	Properly pass down MS_INVALIDATE to lower levels of the VM code from
	vm_map_clean.
15)	Better support the notion of pages being busy but valid, so that
	fewer in-transit waits occur.  (use p->busy more for pageouts instead
	of PG_BUSY.)  Since the page is fully valid, it is still usable for
	reads.
16)	It is possible (in error) for cached pages to be busy.  Make the
	page allocation code handle that case correctly.  (It should probably
	be a printf or panic, but I want the system to handle coding errors
	robustly.  I'll probably add a printf.)
17)	Correct the design and usage of vm_page_sleep.  It didn't handle
	consistancy problems very well, so make the design a little less
	lofty.  After vm_page_sleep, if it ever blocked, it is still important
	to relookup the page (if the object generation count changed), and
	verify it's status (always.)
18)	In vm_pageout.c, vm_pageout_clean had rotted, so clean that up.
19)	Push the page busy for writes and VM_PROT_READ into vm_pageout_flush.
20)	Fix vm_pager_put_pages and it's descendents to support an int flag
	instead of a boolean, so that we can pass down the invalidate bit.
1998-03-07 21:37:31 +00:00

483 lines
12 KiB
C

/*
* Copyright (c) 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed
* to Berkeley by John Heidemann of the UCLA Ficus project.
*
* Source: * @(#)i405_init.c 2.10 92/04/27 UCLA Ficus project
*
* 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.
*
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/unistd.h>
#include <sys/vnode.h>
#include <sys/mount.h>
#include <sys/poll.h>
/*
* VFS operations
*/
/*
* Complement to all vpp returning ops.
* XXX - initially only to get rid of WILLRELE.
*/
/* ARGSUSED */
int
vfs_vrele(mp, vp)
struct mount *mp;
struct vnode *vp;
{
vrele(vp);
return (0);
}
/*
* vnode operations
*/
static int vop_nostrategy __P((struct vop_strategy_args *));
/*
* This vnode table stores what we want to do if the filesystem doesn't
* implement a particular VOP.
*
* If there is no specific entry here, we will return EOPNOTSUPP.
*
*/
vop_t **default_vnodeop_p;
static struct vnodeopv_entry_desc default_vnodeop_entries[] = {
{ &vop_default_desc, (vop_t *) vop_eopnotsupp },
{ &vop_abortop_desc, (vop_t *) vop_null },
{ &vop_advlock_desc, (vop_t *) vop_einval },
{ &vop_bwrite_desc, (vop_t *) vop_stdbwrite },
{ &vop_close_desc, (vop_t *) vop_null },
{ &vop_fsync_desc, (vop_t *) vop_null },
{ &vop_ioctl_desc, (vop_t *) vop_enotty },
{ &vop_islocked_desc, (vop_t *) vop_noislocked },
{ &vop_lease_desc, (vop_t *) vop_null },
{ &vop_lock_desc, (vop_t *) vop_nolock },
{ &vop_mmap_desc, (vop_t *) vop_einval },
{ &vop_open_desc, (vop_t *) vop_null },
{ &vop_pathconf_desc, (vop_t *) vop_einval },
{ &vop_poll_desc, (vop_t *) vop_nopoll },
{ &vop_readlink_desc, (vop_t *) vop_einval },
{ &vop_reallocblks_desc, (vop_t *) vop_eopnotsupp },
{ &vop_revoke_desc, (vop_t *) vop_revoke },
{ &vop_strategy_desc, (vop_t *) vop_nostrategy },
{ &vop_unlock_desc, (vop_t *) vop_nounlock },
{ NULL, NULL }
};
static struct vnodeopv_desc default_vnodeop_opv_desc =
{ &default_vnodeop_p, default_vnodeop_entries };
VNODEOP_SET(default_vnodeop_opv_desc);
int
vop_eopnotsupp(struct vop_generic_args *ap)
{
/*
printf("vop_notsupp[%s]\n", ap->a_desc->vdesc_name);
*/
return (EOPNOTSUPP);
}
int
vop_ebadf(struct vop_generic_args *ap)
{
return (EBADF);
}
int
vop_enotty(struct vop_generic_args *ap)
{
return (ENOTTY);
}
int
vop_einval(struct vop_generic_args *ap)
{
return (EINVAL);
}
int
vop_null(struct vop_generic_args *ap)
{
return (0);
}
int
vop_defaultop(struct vop_generic_args *ap)
{
return (VOCALL(default_vnodeop_p, ap->a_desc->vdesc_offset, ap));
}
static int
vop_nostrategy (struct vop_strategy_args *ap)
{
printf("No strategy for buffer at %p\n", ap->a_bp);
vprint("", ap->a_bp->b_vp);
ap->a_bp->b_flags |= B_ERROR;
ap->a_bp->b_error = EOPNOTSUPP;
biodone(ap->a_bp);
return (EOPNOTSUPP);
}
int
vop_stdpathconf(ap)
struct vop_pathconf_args /* {
struct vnode *a_vp;
int a_name;
int *a_retval;
} */ *ap;
{
switch (ap->a_name) {
case _PC_LINK_MAX:
*ap->a_retval = LINK_MAX;
return (0);
case _PC_MAX_CANON:
*ap->a_retval = MAX_CANON;
return (0);
case _PC_MAX_INPUT:
*ap->a_retval = MAX_INPUT;
return (0);
case _PC_PIPE_BUF:
*ap->a_retval = PIPE_BUF;
return (0);
case _PC_CHOWN_RESTRICTED:
*ap->a_retval = 1;
return (0);
case _PC_VDISABLE:
*ap->a_retval = _POSIX_VDISABLE;
return (0);
default:
return (EINVAL);
}
/* NOTREACHED */
}
/*
* Standard lock, unlock and islocked functions.
*
* These depend on the lock structure being the first element in the
* inode, ie: vp->v_data points to the the lock!
*/
int
vop_stdlock(ap)
struct vop_lock_args /* {
struct vnode *a_vp;
int a_flags;
struct proc *a_p;
} */ *ap;
{
struct lock *l;
if ((l = (struct lock *)ap->a_vp->v_data) == NULL) {
if (ap->a_flags & LK_INTERLOCK)
simple_unlock(&ap->a_vp->v_interlock);
return 0;
}
return (lockmgr(l, ap->a_flags, &ap->a_vp->v_interlock, ap->a_p));
}
int
vop_stdunlock(ap)
struct vop_unlock_args /* {
struct vnode *a_vp;
int a_flags;
struct proc *a_p;
} */ *ap;
{
struct lock *l;
if ((l = (struct lock *)ap->a_vp->v_data) == NULL) {
if (ap->a_flags & LK_INTERLOCK)
simple_unlock(&ap->a_vp->v_interlock);
return 0;
}
return (lockmgr(l, ap->a_flags | LK_RELEASE, &ap->a_vp->v_interlock,
ap->a_p));
}
int
vop_stdislocked(ap)
struct vop_islocked_args /* {
struct vnode *a_vp;
} */ *ap;
{
struct lock *l;
if ((l = (struct lock *)ap->a_vp->v_data) == NULL)
return 0;
return (lockstatus(l));
}
/*
* Return true for select/poll.
*/
int
vop_nopoll(ap)
struct vop_poll_args /* {
struct vnode *a_vp;
int a_events;
struct ucred *a_cred;
struct proc *a_p;
} */ *ap;
{
/*
* Return true for read/write. If the user asked for something
* special, return POLLNVAL, so that clients have a way of
* determining reliably whether or not the extended
* functionality is present without hard-coding knowledge
* of specific filesystem implementations.
*/
if (ap->a_events & ~POLLSTANDARD)
return (POLLNVAL);
return (ap->a_events & (POLLIN | POLLOUT | POLLRDNORM | POLLWRNORM));
}
/*
* Implement poll for local filesystems that support it.
*/
int
vop_stdpoll(ap)
struct vop_poll_args /* {
struct vnode *a_vp;
int a_events;
struct ucred *a_cred;
struct proc *a_p;
} */ *ap;
{
if ((ap->a_events & ~POLLSTANDARD) == 0)
return (ap->a_events & (POLLRDNORM|POLLWRNORM));
return (vn_pollrecord(ap->a_vp, ap->a_p, ap->a_events));
}
int
vop_stdbwrite(ap)
struct vop_bwrite_args *ap;
{
return (bwrite(ap->a_bp));
}
/*
* Stubs to use when there is no locking to be done on the underlying object.
* A minimal shared lock is necessary to ensure that the underlying object
* is not revoked while an operation is in progress. So, an active shared
* count is maintained in an auxillary vnode lock structure.
*/
int
vop_sharedlock(ap)
struct vop_lock_args /* {
struct vnode *a_vp;
int a_flags;
struct proc *a_p;
} */ *ap;
{
/*
* This code cannot be used until all the non-locking filesystems
* (notably NFS) are converted to properly lock and release nodes.
* Also, certain vnode operations change the locking state within
* the operation (create, mknod, remove, link, rename, mkdir, rmdir,
* and symlink). Ideally these operations should not change the
* lock state, but should be changed to let the caller of the
* function unlock them. Otherwise all intermediate vnode layers
* (such as union, umapfs, etc) must catch these functions to do
* the necessary locking at their layer. Note that the inactive
* and lookup operations also change their lock state, but this
* cannot be avoided, so these two operations will always need
* to be handled in intermediate layers.
*/
struct vnode *vp = ap->a_vp;
int vnflags, flags = ap->a_flags;
if (vp->v_vnlock == NULL) {
if ((flags & LK_TYPE_MASK) == LK_DRAIN)
return (0);
MALLOC(vp->v_vnlock, struct lock *, sizeof(struct lock),
M_VNODE, M_WAITOK);
lockinit(vp->v_vnlock, PVFS, "vnlock", 0, LK_NOPAUSE);
}
switch (flags & LK_TYPE_MASK) {
case LK_DRAIN:
vnflags = LK_DRAIN;
break;
case LK_EXCLUSIVE:
#ifdef DEBUG_VFS_LOCKS
/*
* Normally, we use shared locks here, but that confuses
* the locking assertions.
*/
vnflags = LK_EXCLUSIVE;
break;
#endif
case LK_SHARED:
vnflags = LK_SHARED;
break;
case LK_UPGRADE:
case LK_EXCLUPGRADE:
case LK_DOWNGRADE:
return (0);
case LK_RELEASE:
default:
panic("vop_sharedlock: bad operation %d", flags & LK_TYPE_MASK);
}
if (flags & LK_INTERLOCK)
vnflags |= LK_INTERLOCK;
return(lockmgr(vp->v_vnlock, vnflags, &vp->v_interlock, ap->a_p));
}
/*
* Stubs to use when there is no locking to be done on the underlying object.
* A minimal shared lock is necessary to ensure that the underlying object
* is not revoked while an operation is in progress. So, an active shared
* count is maintained in an auxillary vnode lock structure.
*/
int
vop_nolock(ap)
struct vop_lock_args /* {
struct vnode *a_vp;
int a_flags;
struct proc *a_p;
} */ *ap;
{
#ifdef notyet
/*
* This code cannot be used until all the non-locking filesystems
* (notably NFS) are converted to properly lock and release nodes.
* Also, certain vnode operations change the locking state within
* the operation (create, mknod, remove, link, rename, mkdir, rmdir,
* and symlink). Ideally these operations should not change the
* lock state, but should be changed to let the caller of the
* function unlock them. Otherwise all intermediate vnode layers
* (such as union, umapfs, etc) must catch these functions to do
* the necessary locking at their layer. Note that the inactive
* and lookup operations also change their lock state, but this
* cannot be avoided, so these two operations will always need
* to be handled in intermediate layers.
*/
struct vnode *vp = ap->a_vp;
int vnflags, flags = ap->a_flags;
if (vp->v_vnlock == NULL) {
if ((flags & LK_TYPE_MASK) == LK_DRAIN)
return (0);
MALLOC(vp->v_vnlock, struct lock *, sizeof(struct lock),
M_VNODE, M_WAITOK);
lockinit(vp->v_vnlock, PVFS, "vnlock", 0, LK_NOPAUSE);
}
switch (flags & LK_TYPE_MASK) {
case LK_DRAIN:
vnflags = LK_DRAIN;
break;
case LK_EXCLUSIVE:
case LK_SHARED:
vnflags = LK_SHARED;
break;
case LK_UPGRADE:
case LK_EXCLUPGRADE:
case LK_DOWNGRADE:
return (0);
case LK_RELEASE:
default:
panic("vop_nolock: bad operation %d", flags & LK_TYPE_MASK);
}
if (flags & LK_INTERLOCK)
vnflags |= LK_INTERLOCK;
return(lockmgr(vp->v_vnlock, vnflags, &vp->v_interlock, ap->a_p));
#else /* for now */
/*
* Since we are not using the lock manager, we must clear
* the interlock here.
*/
if (ap->a_flags & LK_INTERLOCK)
simple_unlock(&ap->a_vp->v_interlock);
return (0);
#endif
}
/*
* Do the inverse of vop_nolock, handling the interlock in a compatible way.
*/
int
vop_nounlock(ap)
struct vop_unlock_args /* {
struct vnode *a_vp;
int a_flags;
struct proc *a_p;
} */ *ap;
{
struct vnode *vp = ap->a_vp;
if (vp->v_vnlock == NULL) {
if (ap->a_flags & LK_INTERLOCK)
simple_unlock(&ap->a_vp->v_interlock);
return (0);
}
return (lockmgr(vp->v_vnlock, LK_RELEASE | ap->a_flags,
&ap->a_vp->v_interlock, ap->a_p));
}
/*
* Return whether or not the node is in use.
*/
int
vop_noislocked(ap)
struct vop_islocked_args /* {
struct vnode *a_vp;
} */ *ap;
{
struct vnode *vp = ap->a_vp;
if (vp->v_vnlock == NULL)
return (0);
return (lockstatus(vp->v_vnlock));
}