HardenedBSD/sys/vm/kern_lock.c
David Greenman 0d94caffca These changes embody the support of the fully coherent merged VM buffer cache,
much higher filesystem I/O performance, and much better paging performance. It
represents the culmination of over 6 months of R&D.

The majority of the merged VM/cache work is by John Dyson.

The following highlights the most significant changes. Additionally, there are
(mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to
support the new VM/buffer scheme.

vfs_bio.c:
Significant rewrite of most of vfs_bio to support the merged VM buffer cache
scheme.  The scheme is almost fully compatible with the old filesystem
interface.  Significant improvement in the number of opportunities for write
clustering.

vfs_cluster.c, vfs_subr.c
Upgrade and performance enhancements in vfs layer code to support merged
VM/buffer cache.  Fixup of vfs_cluster to eliminate the bogus pagemove stuff.

vm_object.c:
Yet more improvements in the collapse code.  Elimination of some windows that
can cause list corruption.

vm_pageout.c:
Fixed it, it really works better now.  Somehow in 2.0, some "enhancements"
broke the code.  This code has been reworked from the ground-up.

vm_fault.c, vm_page.c, pmap.c, vm_object.c
Support for small-block filesystems with merged VM/buffer cache scheme.

pmap.c vm_map.c
Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of
kernel PTs.

vm_glue.c
Much simpler and more effective swapping code.  No more gratuitous swapping.

proc.h
Fixed the problem that the p_lock flag was not being cleared on a fork.

swap_pager.c, vnode_pager.c
Removal of old vfs_bio cruft to support the past pseudo-coherency.  Now the
code doesn't need it anymore.

machdep.c
Changes to better support the parameter values for the merged VM/buffer cache
scheme.

machdep.c, kern_exec.c, vm_glue.c
Implemented a seperate submap for temporary exec string space and another one
to contain process upages. This eliminates all map fragmentation problems
that previously existed.

ffs_inode.c, ufs_inode.c, ufs_readwrite.c
Changes for merged VM/buffer cache.  Add "bypass" support for sneaking in on
busy buffers.

Submitted by:	John Dyson and David Greenman
1995-01-09 16:06:02 +00:00

536 lines
12 KiB
C

/*
* Copyright (c) 1991, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* The Mach Operating System project at Carnegie-Mellon University.
*
* 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.
*
* from: @(#)kern_lock.c 8.1 (Berkeley) 6/11/93
*
*
* Copyright (c) 1987, 1990 Carnegie-Mellon University.
* All rights reserved.
*
* Authors: Avadis Tevanian, Jr., Michael Wayne Young
*
* Permission to use, copy, modify and distribute this software and
* its documentation is hereby granted, provided that both the copyright
* notice and this permission notice appear in all copies of the
* software, derivative works or modified versions, and any portions
* thereof, and that both notices appear in supporting documentation.
*
* CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
* CONDITION. CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND
* FOR ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
*
* Carnegie Mellon requests users of this software to return to
*
* Software Distribution Coordinator or Software.Distribution@CS.CMU.EDU
* School of Computer Science
* Carnegie Mellon University
* Pittsburgh PA 15213-3890
*
* any improvements or extensions that they make and grant Carnegie the
* rights to redistribute these changes.
*
* $Id: kern_lock.c,v 1.2 1994/08/02 07:55:08 davidg Exp $
*/
/*
* Locking primitives implementation
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <vm/vm.h>
/* XXX */
#include <sys/proc.h>
typedef int *thread_t;
#define current_thread() ((thread_t)&curproc->p_thread)
/* XXX */
#if NCPUS > 1
/*
* Module: lock
* Function:
* Provide reader/writer sychronization.
* Implementation:
* Simple interlock on a bit. Readers first interlock
* increment the reader count, then let go. Writers hold
* the interlock (thus preventing further readers), and
* wait for already-accepted readers to go away.
*/
/*
* The simple-lock routines are the primitives out of which
* the lock package is built. The implementation is left
* to the machine-dependent code.
*/
#ifdef notdef
/*
* A sample implementation of simple locks.
* assumes:
* boolean_t test_and_set(boolean_t *)
* indivisibly sets the boolean to TRUE
* and returns its old value
* and that setting a boolean to FALSE is indivisible.
*/
/*
* simple_lock_init initializes a simple lock. A simple lock
* may only be used for exclusive locks.
*/
void
simple_lock_init(l)
simple_lock_t l;
{
*(boolean_t *) l = FALSE;
}
void
simple_lock(l)
simple_lock_t l;
{
while (test_and_set((boolean_t *) l))
continue;
}
void
simple_unlock(l)
simple_lock_t l;
{
*(boolean_t *) l = FALSE;
}
boolean_t
simple_lock_try(l)
simple_lock_t l;
{
return (!test_and_set((boolean_t *) l));
}
#endif /* notdef */
#endif /* NCPUS > 1 */
#if NCPUS > 1
int lock_wait_time = 100;
#else /* NCPUS > 1 */
/*
* It is silly to spin on a uni-processor as if we thought something magical
* would happen to the want_write bit while we are executing.
*/
int lock_wait_time = 0;
#endif /* NCPUS > 1 */
/*
* Routine: lock_init
* Function:
* Initialize a lock; required before use.
* Note that clients declare the "struct lock"
* variables and then initialize them, rather
* than getting a new one from this module.
*/
void
lock_init(l, can_sleep)
lock_t l;
boolean_t can_sleep;
{
bzero(l, sizeof(lock_data_t));
simple_lock_init(&l->interlock);
l->want_write = FALSE;
l->want_upgrade = FALSE;
l->read_count = 0;
l->can_sleep = can_sleep;
l->thread = (char *) -1; /* XXX */
l->recursion_depth = 0;
}
void
lock_sleepable(l, can_sleep)
lock_t l;
boolean_t can_sleep;
{
simple_lock(&l->interlock);
l->can_sleep = can_sleep;
simple_unlock(&l->interlock);
}
/*
* Sleep locks. These use the same data structure and algorithm
* as the spin locks, but the process sleeps while it is waiting
* for the lock. These work on uniprocessor systems.
*/
void
lock_write(l)
register lock_t l;
{
register int i;
simple_lock(&l->interlock);
if (((thread_t) l->thread) == current_thread()) {
/*
* Recursive lock.
*/
l->recursion_depth++;
simple_unlock(&l->interlock);
return;
}
/*
* Try to acquire the want_write bit.
*/
while (l->want_write) {
if ((i = lock_wait_time) > 0) {
simple_unlock(&l->interlock);
while (--i > 0 && l->want_write)
continue;
simple_lock(&l->interlock);
}
if (l->can_sleep && l->want_write) {
l->waiting = TRUE;
thread_sleep((int) l, &l->interlock, FALSE);
simple_lock(&l->interlock);
}
}
l->want_write = TRUE;
/* Wait for readers (and upgrades) to finish */
while ((l->read_count != 0) || l->want_upgrade) {
if ((i = lock_wait_time) > 0) {
simple_unlock(&l->interlock);
while (--i > 0 && (l->read_count != 0 ||
l->want_upgrade))
continue;
simple_lock(&l->interlock);
}
if (l->can_sleep && (l->read_count != 0 || l->want_upgrade)) {
l->waiting = TRUE;
thread_sleep((int) l, &l->interlock, FALSE);
simple_lock(&l->interlock);
}
}
simple_unlock(&l->interlock);
}
void
lock_done(l)
register lock_t l;
{
simple_lock(&l->interlock);
if (l->read_count != 0)
l->read_count--;
else if (l->recursion_depth != 0)
l->recursion_depth--;
else if (l->want_upgrade)
l->want_upgrade = FALSE;
else
l->want_write = FALSE;
if (l->waiting) {
l->waiting = FALSE;
thread_wakeup((int) l);
}
simple_unlock(&l->interlock);
}
void
lock_read(l)
register lock_t l;
{
register int i;
simple_lock(&l->interlock);
if (((thread_t) l->thread) == current_thread()) {
/*
* Recursive lock.
*/
l->read_count++;
simple_unlock(&l->interlock);
return;
}
while (l->want_write || l->want_upgrade) {
if ((i = lock_wait_time) > 0) {
simple_unlock(&l->interlock);
while (--i > 0 && (l->want_write || l->want_upgrade))
continue;
simple_lock(&l->interlock);
}
if (l->can_sleep && (l->want_write || l->want_upgrade)) {
l->waiting = TRUE;
thread_sleep((int) l, &l->interlock, FALSE);
simple_lock(&l->interlock);
}
}
l->read_count++;
simple_unlock(&l->interlock);
}
/*
* Routine: lock_read_to_write
* Function:
* Improves a read-only lock to one with
* write permission. If another reader has
* already requested an upgrade to a write lock,
* no lock is held upon return.
*
* Returns TRUE if the upgrade *failed*.
*/
boolean_t
lock_read_to_write(l)
register lock_t l;
{
register int i;
simple_lock(&l->interlock);
l->read_count--;
if (((thread_t) l->thread) == current_thread()) {
/*
* Recursive lock.
*/
l->recursion_depth++;
simple_unlock(&l->interlock);
return (FALSE);
}
if (l->want_upgrade) {
/*
* Someone else has requested upgrade. Since we've released a
* read lock, wake him up.
*/
if (l->waiting) {
l->waiting = FALSE;
thread_wakeup((int) l);
}
simple_unlock(&l->interlock);
return (TRUE);
}
l->want_upgrade = TRUE;
while (l->read_count != 0) {
if ((i = lock_wait_time) > 0) {
simple_unlock(&l->interlock);
while (--i > 0 && l->read_count != 0)
continue;
simple_lock(&l->interlock);
}
if (l->can_sleep && l->read_count != 0) {
l->waiting = TRUE;
thread_sleep((int) l, &l->interlock, FALSE);
simple_lock(&l->interlock);
}
}
simple_unlock(&l->interlock);
return (FALSE);
}
void
lock_write_to_read(l)
register lock_t l;
{
simple_lock(&l->interlock);
l->read_count++;
if (l->recursion_depth != 0)
l->recursion_depth--;
else if (l->want_upgrade)
l->want_upgrade = FALSE;
else
l->want_write = FALSE;
if (l->waiting) {
l->waiting = FALSE;
thread_wakeup((int) l);
}
simple_unlock(&l->interlock);
}
/*
* Routine: lock_try_write
* Function:
* Tries to get a write lock.
*
* Returns FALSE if the lock is not held on return.
*/
boolean_t
lock_try_write(l)
register lock_t l;
{
simple_lock(&l->interlock);
if (((thread_t) l->thread) == current_thread()) {
/*
* Recursive lock
*/
l->recursion_depth++;
simple_unlock(&l->interlock);
return (TRUE);
}
if (l->want_write || l->want_upgrade || l->read_count) {
/*
* Can't get lock.
*/
simple_unlock(&l->interlock);
return (FALSE);
}
/*
* Have lock.
*/
l->want_write = TRUE;
simple_unlock(&l->interlock);
return (TRUE);
}
/*
* Routine: lock_try_read
* Function:
* Tries to get a read lock.
*
* Returns FALSE if the lock is not held on return.
*/
boolean_t
lock_try_read(l)
register lock_t l;
{
simple_lock(&l->interlock);
if (((thread_t) l->thread) == current_thread()) {
/*
* Recursive lock
*/
l->read_count++;
simple_unlock(&l->interlock);
return (TRUE);
}
if (l->want_write || l->want_upgrade) {
simple_unlock(&l->interlock);
return (FALSE);
}
l->read_count++;
simple_unlock(&l->interlock);
return (TRUE);
}
/*
* Routine: lock_try_read_to_write
* Function:
* Improves a read-only lock to one with
* write permission. If another reader has
* already requested an upgrade to a write lock,
* the read lock is still held upon return.
*
* Returns FALSE if the upgrade *failed*.
*/
boolean_t
lock_try_read_to_write(l)
register lock_t l;
{
simple_lock(&l->interlock);
if (((thread_t) l->thread) == current_thread()) {
/*
* Recursive lock
*/
l->read_count--;
l->recursion_depth++;
simple_unlock(&l->interlock);
return (TRUE);
}
if (l->want_upgrade) {
simple_unlock(&l->interlock);
return (FALSE);
}
l->want_upgrade = TRUE;
l->read_count--;
while (l->read_count != 0) {
l->waiting = TRUE;
thread_sleep((int) l, &l->interlock, FALSE);
simple_lock(&l->interlock);
}
simple_unlock(&l->interlock);
return (TRUE);
}
/*
* Allow a process that has a lock for write to acquire it
* recursively (for read, write, or update).
*/
void
lock_set_recursive(l)
lock_t l;
{
simple_lock(&l->interlock);
if (!l->want_write) {
panic("lock_set_recursive: don't have write lock");
}
l->thread = (char *) current_thread();
simple_unlock(&l->interlock);
}
/*
* Prevent a lock from being re-acquired.
*/
void
lock_clear_recursive(l)
lock_t l;
{
simple_lock(&l->interlock);
if (((thread_t) l->thread) != current_thread()) {
panic("lock_clear_recursive: wrong thread");
}
if (l->recursion_depth == 0)
l->thread = (char *) -1; /* XXX */
simple_unlock(&l->interlock);
}