2004-05-13 00:09:50 +04:00
|
|
|
/* $NetBSD: uvm_page_i.h,v 1.22 2004/05/12 20:09:52 yamt Exp $ */
|
1998-02-05 09:25:08 +03:00
|
|
|
|
2001-05-25 08:06:11 +04:00
|
|
|
/*
|
1998-02-05 09:25:08 +03:00
|
|
|
* Copyright (c) 1997 Charles D. Cranor and Washington University.
|
2001-05-25 08:06:11 +04:00
|
|
|
* Copyright (c) 1991, 1993, The Regents of the University of California.
|
1998-02-05 09:25:08 +03:00
|
|
|
*
|
|
|
|
* 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 Charles D. Cranor,
|
2001-05-25 08:06:11 +04:00
|
|
|
* Washington University, the University of California, Berkeley and
|
1998-02-05 09:25:08 +03:00
|
|
|
* 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.
|
|
|
|
*
|
|
|
|
* @(#)vm_page.c 8.3 (Berkeley) 3/21/94
|
1998-02-07 14:07:38 +03:00
|
|
|
* from: Id: uvm_page_i.h,v 1.1.2.7 1998/01/05 00:26:02 chuck Exp
|
1998-02-05 09:25:08 +03:00
|
|
|
*
|
|
|
|
*
|
|
|
|
* Copyright (c) 1987, 1990 Carnegie-Mellon University.
|
|
|
|
* All rights reserved.
|
2001-05-25 08:06:11 +04:00
|
|
|
*
|
1998-02-05 09:25:08 +03:00
|
|
|
* 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.
|
2001-05-25 08:06:11 +04:00
|
|
|
*
|
|
|
|
* CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
|
|
|
|
* CONDITION. CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND
|
1998-02-05 09:25:08 +03:00
|
|
|
* FOR ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
|
2001-05-25 08:06:11 +04:00
|
|
|
*
|
1998-02-05 09:25:08 +03:00
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uvm_page_i.h
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* inline functions [maybe]
|
|
|
|
*/
|
|
|
|
|
|
|
|
#if defined(UVM_PAGE_INLINE) || defined(UVM_PAGE)
|
|
|
|
|
2002-12-02 01:58:43 +03:00
|
|
|
#ifndef _UVM_UVM_PAGE_I_H_
|
|
|
|
#define _UVM_UVM_PAGE_I_H_
|
|
|
|
|
1999-05-24 23:10:57 +04:00
|
|
|
/*
|
|
|
|
* uvm_lock_fpageq: lock the free page queue
|
|
|
|
*
|
|
|
|
* => free page queue can be accessed in interrupt context, so this
|
|
|
|
* blocks all interrupts that can cause memory allocation, and
|
|
|
|
* returns the previous interrupt level.
|
|
|
|
*/
|
|
|
|
|
|
|
|
PAGE_INLINE int
|
|
|
|
uvm_lock_fpageq()
|
|
|
|
{
|
|
|
|
int s;
|
|
|
|
|
2001-01-14 05:10:01 +03:00
|
|
|
s = splvm();
|
1999-05-24 23:10:57 +04:00
|
|
|
simple_lock(&uvm.fpageqlock);
|
|
|
|
return (s);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uvm_unlock_fpageq: unlock the free page queue
|
|
|
|
*
|
|
|
|
* => caller must supply interrupt level returned by uvm_lock_fpageq()
|
|
|
|
* so that it may be restored.
|
|
|
|
*/
|
|
|
|
|
|
|
|
PAGE_INLINE void
|
|
|
|
uvm_unlock_fpageq(s)
|
|
|
|
int s;
|
|
|
|
{
|
|
|
|
|
|
|
|
simple_unlock(&uvm.fpageqlock);
|
|
|
|
splx(s);
|
|
|
|
}
|
|
|
|
|
1998-02-05 09:25:08 +03:00
|
|
|
/*
|
|
|
|
* uvm_pagelookup: look up a page
|
|
|
|
*
|
|
|
|
* => caller should lock object to keep someone from pulling the page
|
|
|
|
* out from under it
|
|
|
|
*/
|
|
|
|
|
1998-03-09 03:58:55 +03:00
|
|
|
struct vm_page *
|
|
|
|
uvm_pagelookup(obj, off)
|
|
|
|
struct uvm_object *obj;
|
2000-03-27 00:54:45 +04:00
|
|
|
voff_t off;
|
1998-02-05 09:25:08 +03:00
|
|
|
{
|
1998-03-09 03:58:55 +03:00
|
|
|
struct vm_page *pg;
|
|
|
|
struct pglist *buck;
|
|
|
|
|
|
|
|
buck = &uvm.page_hash[uvm_pagehash(obj,off)];
|
|
|
|
simple_lock(&uvm.hashlock);
|
2000-11-27 10:47:42 +03:00
|
|
|
TAILQ_FOREACH(pg, buck, hashq) {
|
1998-03-09 03:58:55 +03:00
|
|
|
if (pg->uobject == obj && pg->offset == off) {
|
2000-11-27 10:47:42 +03:00
|
|
|
break;
|
1998-03-09 03:58:55 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
simple_unlock(&uvm.hashlock);
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
KASSERT(pg == NULL || obj->uo_npages != 0);
|
|
|
|
KASSERT(pg == NULL || (pg->flags & (PG_RELEASED|PG_PAGEOUT)) == 0 ||
|
|
|
|
(pg->flags & PG_BUSY) != 0);
|
2000-11-27 10:47:42 +03:00
|
|
|
return(pg);
|
1998-02-05 09:25:08 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
1998-03-23 00:29:30 +03:00
|
|
|
* uvm_pagewire: wire the page, thus removing it from the daemon's grasp
|
1998-02-05 09:25:08 +03:00
|
|
|
*
|
|
|
|
* => caller must lock page queues
|
|
|
|
*/
|
|
|
|
|
1998-03-09 03:58:55 +03:00
|
|
|
PAGE_INLINE void
|
1998-03-23 00:29:30 +03:00
|
|
|
uvm_pagewire(pg)
|
1998-03-09 03:58:55 +03:00
|
|
|
struct vm_page *pg;
|
1998-02-05 09:25:08 +03:00
|
|
|
{
|
2004-05-13 00:09:50 +04:00
|
|
|
UVM_LOCK_ASSERT_PAGEQ();
|
1998-03-09 03:58:55 +03:00
|
|
|
if (pg->wire_count == 0) {
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
uvm_pagedequeue(pg);
|
1998-03-09 03:58:55 +03:00
|
|
|
uvmexp.wired++;
|
|
|
|
}
|
1998-03-23 00:29:30 +03:00
|
|
|
pg->wire_count++;
|
1998-02-05 09:25:08 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2001-05-25 08:06:11 +04:00
|
|
|
* uvm_pageunwire: unwire the page.
|
1998-02-05 09:25:08 +03:00
|
|
|
*
|
|
|
|
* => activate if wire count goes to zero.
|
|
|
|
* => caller must lock page queues
|
|
|
|
*/
|
2001-05-25 08:06:11 +04:00
|
|
|
|
1998-03-09 03:58:55 +03:00
|
|
|
PAGE_INLINE void
|
|
|
|
uvm_pageunwire(pg)
|
|
|
|
struct vm_page *pg;
|
1998-02-05 09:25:08 +03:00
|
|
|
{
|
2004-05-13 00:09:50 +04:00
|
|
|
UVM_LOCK_ASSERT_PAGEQ();
|
1998-03-09 03:58:55 +03:00
|
|
|
pg->wire_count--;
|
|
|
|
if (pg->wire_count == 0) {
|
|
|
|
TAILQ_INSERT_TAIL(&uvm.page_active, pg, pageq);
|
|
|
|
uvmexp.active++;
|
|
|
|
pg->pqflags |= PQ_ACTIVE;
|
|
|
|
uvmexp.wired--;
|
|
|
|
}
|
1998-02-05 09:25:08 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
Page scanner improvements, behavior is actually a bit more like
Mach VM's now. Specific changes:
- Pages now need not have all of their mappings removed before being
put on the inactive list. They only need to have the "referenced"
attribute cleared. This makes putting pages onto the inactive list
much more efficient. In order to eliminate redundant clearings of
"refrenced", callers of uvm_pagedeactivate() must now do this
themselves.
- When checking the "modified" attribute for a page (for clearing
PG_CLEAN), make sure to only do it if PG_CLEAN is currently set on
the page (saves a potentially expensive pmap operation).
- When scanning the inactive list, if a page is referenced, reactivate
it (this part was actually added in uvm_pdaemon.c,v 1.27). This
now works properly now that pages on the inactive list are allowed to
have mappings.
- When scanning the inactive list and considering a page for freeing,
remove all mappings, and then check the "modified" attribute if the
page is marked PG_CLEAN.
- When scanning the active list, if the page was referenced since its
last sweep by the scanner, don't deactivate it. (This part was
actually added in uvm_pdaemon.c,v 1.28.)
These changes greatly improve interactive performance during
moderate to high memory and I/O load.
2001-01-29 02:30:42 +03:00
|
|
|
* uvm_pagedeactivate: deactivate page
|
1998-02-05 09:25:08 +03:00
|
|
|
*
|
|
|
|
* => caller must lock page queues
|
|
|
|
* => caller must check to make sure page is not wired
|
|
|
|
* => object that page belongs to must be locked (so we can adjust pg->flags)
|
Page scanner improvements, behavior is actually a bit more like
Mach VM's now. Specific changes:
- Pages now need not have all of their mappings removed before being
put on the inactive list. They only need to have the "referenced"
attribute cleared. This makes putting pages onto the inactive list
much more efficient. In order to eliminate redundant clearings of
"refrenced", callers of uvm_pagedeactivate() must now do this
themselves.
- When checking the "modified" attribute for a page (for clearing
PG_CLEAN), make sure to only do it if PG_CLEAN is currently set on
the page (saves a potentially expensive pmap operation).
- When scanning the inactive list, if a page is referenced, reactivate
it (this part was actually added in uvm_pdaemon.c,v 1.27). This
now works properly now that pages on the inactive list are allowed to
have mappings.
- When scanning the inactive list and considering a page for freeing,
remove all mappings, and then check the "modified" attribute if the
page is marked PG_CLEAN.
- When scanning the active list, if the page was referenced since its
last sweep by the scanner, don't deactivate it. (This part was
actually added in uvm_pdaemon.c,v 1.28.)
These changes greatly improve interactive performance during
moderate to high memory and I/O load.
2001-01-29 02:30:42 +03:00
|
|
|
* => caller must clear the reference on the page before calling
|
1998-02-05 09:25:08 +03:00
|
|
|
*/
|
|
|
|
|
1998-03-09 03:58:55 +03:00
|
|
|
PAGE_INLINE void
|
|
|
|
uvm_pagedeactivate(pg)
|
|
|
|
struct vm_page *pg;
|
1998-02-05 09:25:08 +03:00
|
|
|
{
|
2004-05-13 00:09:50 +04:00
|
|
|
UVM_LOCK_ASSERT_PAGEQ();
|
1998-03-09 03:58:55 +03:00
|
|
|
if (pg->pqflags & PQ_ACTIVE) {
|
|
|
|
TAILQ_REMOVE(&uvm.page_active, pg, pageq);
|
|
|
|
pg->pqflags &= ~PQ_ACTIVE;
|
|
|
|
uvmexp.active--;
|
|
|
|
}
|
|
|
|
if ((pg->pqflags & PQ_INACTIVE) == 0) {
|
2000-11-27 10:47:42 +03:00
|
|
|
KASSERT(pg->wire_count == 0);
|
2001-05-22 04:44:44 +04:00
|
|
|
TAILQ_INSERT_TAIL(&uvm.page_inactive, pg, pageq);
|
1998-03-09 03:58:55 +03:00
|
|
|
pg->pqflags |= PQ_INACTIVE;
|
|
|
|
uvmexp.inactive++;
|
|
|
|
}
|
1998-02-05 09:25:08 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uvm_pageactivate: activate page
|
|
|
|
*
|
|
|
|
* => caller must lock page queues
|
|
|
|
*/
|
|
|
|
|
1998-03-09 03:58:55 +03:00
|
|
|
PAGE_INLINE void
|
|
|
|
uvm_pageactivate(pg)
|
|
|
|
struct vm_page *pg;
|
1998-02-05 09:25:08 +03:00
|
|
|
{
|
2004-05-13 00:09:50 +04:00
|
|
|
UVM_LOCK_ASSERT_PAGEQ();
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
uvm_pagedequeue(pg);
|
1998-03-09 03:58:55 +03:00
|
|
|
if (pg->wire_count == 0) {
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
TAILQ_INSERT_TAIL(&uvm.page_active, pg, pageq);
|
|
|
|
pg->pqflags |= PQ_ACTIVE;
|
|
|
|
uvmexp.active++;
|
|
|
|
}
|
|
|
|
}
|
1998-03-09 03:58:55 +03:00
|
|
|
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
/*
|
|
|
|
* uvm_pagedequeue: remove a page from any paging queue
|
|
|
|
*/
|
1998-03-09 03:58:55 +03:00
|
|
|
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
PAGE_INLINE void
|
|
|
|
uvm_pagedequeue(pg)
|
|
|
|
struct vm_page *pg;
|
|
|
|
{
|
|
|
|
if (pg->pqflags & PQ_ACTIVE) {
|
2004-05-13 00:09:50 +04:00
|
|
|
UVM_LOCK_ASSERT_PAGEQ();
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
TAILQ_REMOVE(&uvm.page_active, pg, pageq);
|
|
|
|
pg->pqflags &= ~PQ_ACTIVE;
|
|
|
|
uvmexp.active--;
|
|
|
|
} else if (pg->pqflags & PQ_INACTIVE) {
|
2004-05-13 00:09:50 +04:00
|
|
|
UVM_LOCK_ASSERT_PAGEQ();
|
a whole bunch of changes to improve performance and robustness under load:
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
2001-09-16 00:36:31 +04:00
|
|
|
TAILQ_REMOVE(&uvm.page_inactive, pg, pageq);
|
|
|
|
pg->pqflags &= ~PQ_INACTIVE;
|
|
|
|
uvmexp.inactive--;
|
1998-03-09 03:58:55 +03:00
|
|
|
}
|
1998-02-05 09:25:08 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uvm_pagezero: zero fill a page
|
|
|
|
*
|
|
|
|
* => if page is part of an object then the object should be locked
|
|
|
|
* to protect pg->flags.
|
|
|
|
*/
|
|
|
|
|
1998-03-09 03:58:55 +03:00
|
|
|
PAGE_INLINE void
|
|
|
|
uvm_pagezero(pg)
|
|
|
|
struct vm_page *pg;
|
1998-02-05 09:25:08 +03:00
|
|
|
{
|
1998-03-09 03:58:55 +03:00
|
|
|
pg->flags &= ~PG_CLEAN;
|
|
|
|
pmap_zero_page(VM_PAGE_TO_PHYS(pg));
|
1998-02-05 09:25:08 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uvm_pagecopy: copy a page
|
|
|
|
*
|
|
|
|
* => if page is part of an object then the object should be locked
|
|
|
|
* to protect pg->flags.
|
|
|
|
*/
|
|
|
|
|
1998-03-09 03:58:55 +03:00
|
|
|
PAGE_INLINE void
|
|
|
|
uvm_pagecopy(src, dst)
|
|
|
|
struct vm_page *src, *dst;
|
1998-02-05 09:25:08 +03:00
|
|
|
{
|
1998-03-09 03:58:55 +03:00
|
|
|
|
|
|
|
dst->flags &= ~PG_CLEAN;
|
|
|
|
pmap_copy_page(VM_PAGE_TO_PHYS(src), VM_PAGE_TO_PHYS(dst));
|
1998-02-05 09:25:08 +03:00
|
|
|
}
|
|
|
|
|
Add support for multiple memory free lists. There is at least one
default free list, and 0 - N additional free list, in order of descending
priority.
A new page allocation function, uvm_pagealloc_strat(), has been added,
providing three page allocation strategies:
- normal: high -> low priority free list walk, taking the
page off the first free list that has one.
- only: attempt to allocate a page only from the specified free
list, failing if that free list has none available.
- fallback: if `only' fails, fall back on `normal'.
uvm_pagealloc(...) is provided for normal use (and is a synonym for
uvm_pagealloc_strat(..., UVM_PGA_STRAT_NORMAL, 0); the free list argument
is ignored for the `normal' case).
uvm_page_physload() now specified which free list the pages will be
loaded onto. This means that some platforms which have multiple physical
memory segments may define additional vm_physsegs if they wish to break
individual physical segments into differing priorities.
Machine-dependent code must define _at least_ the following constants
in <machine/vmparam.h>:
VM_NFREELIST: the number of free lists the system will have
VM_FREELIST_DEFAULT: the default freelist (should always be 0,
but is defined in machdep code so that it's with all of the
other free list-related constants).
Additional free list names may be defined by machine-dependent code, but
they will only be used by machine-dependent code (e.g. for loading the
vm_physsegs).
1998-07-08 08:28:27 +04:00
|
|
|
/*
|
|
|
|
* uvm_page_lookup_freelist: look up the free list for the specified page
|
|
|
|
*/
|
|
|
|
|
|
|
|
PAGE_INLINE int
|
|
|
|
uvm_page_lookup_freelist(pg)
|
|
|
|
struct vm_page *pg;
|
|
|
|
{
|
|
|
|
int lcv;
|
|
|
|
|
|
|
|
lcv = vm_physseg_find(atop(VM_PAGE_TO_PHYS(pg)), NULL);
|
2000-11-27 10:47:42 +03:00
|
|
|
KASSERT(lcv != -1);
|
Add support for multiple memory free lists. There is at least one
default free list, and 0 - N additional free list, in order of descending
priority.
A new page allocation function, uvm_pagealloc_strat(), has been added,
providing three page allocation strategies:
- normal: high -> low priority free list walk, taking the
page off the first free list that has one.
- only: attempt to allocate a page only from the specified free
list, failing if that free list has none available.
- fallback: if `only' fails, fall back on `normal'.
uvm_pagealloc(...) is provided for normal use (and is a synonym for
uvm_pagealloc_strat(..., UVM_PGA_STRAT_NORMAL, 0); the free list argument
is ignored for the `normal' case).
uvm_page_physload() now specified which free list the pages will be
loaded onto. This means that some platforms which have multiple physical
memory segments may define additional vm_physsegs if they wish to break
individual physical segments into differing priorities.
Machine-dependent code must define _at least_ the following constants
in <machine/vmparam.h>:
VM_NFREELIST: the number of free lists the system will have
VM_FREELIST_DEFAULT: the default freelist (should always be 0,
but is defined in machdep code so that it's with all of the
other free list-related constants).
Additional free list names may be defined by machine-dependent code, but
they will only be used by machine-dependent code (e.g. for loading the
vm_physsegs).
1998-07-08 08:28:27 +04:00
|
|
|
return (vm_physmem[lcv].free_list);
|
|
|
|
}
|
|
|
|
|
1998-02-10 05:34:17 +03:00
|
|
|
#endif /* _UVM_UVM_PAGE_I_H_ */
|
2002-12-02 01:58:43 +03:00
|
|
|
|
|
|
|
#endif /* defined(UVM_PAGE_INLINE) || defined(UVM_PAGE) */
|